Jump to content




Photo

Stepper motor spec for converting Super 8 projector

scanner super 8 motor DIY

  • Please log in to reply
11 replies to this topic

#1 Simon Lucas

Simon Lucas
  • Basic Members
  • PipPipPip
  • 100 posts
  • Other
  • UK

Posted 30 October 2015 - 10:25 AM

I'm on the cusp of buying the kit I need to convert a Super 8 projector for scanning.

 

I will either cannibalise my nice old Bolex (slightly reluctantly)  or buy a cheap ELMO, or something similar.

 

But assuming that the requirements for the motor will be similar across projectors, I wonder what torque I might need? And I can use a 200 step motor or would I need the more expensive 400 step motor for greater accuracy?

 

Thanks for any advice.


  • 0




#2 Carl Looper

Carl Looper
  • Basic Members
  • PipPipPipPip
  • 1367 posts
  • Digital Image Technician
  • Melbourne, Australia

Posted 19 November 2015 - 01:40 PM

I acquired a cheap force gauge to do this sort of thing. Cost me about $20.

 

Wind a string around the shaft to be measured and then tie the loose end of the string to the gauge. You then pull on the gauge - which effectively measures the force required to get the shaft rotating. Make sure the gauge is set to measure peak force. Resisting your pull on the string will be the load on the shaft. Pulling on the string with the gauge overcomes this resistance, and gives you a reading of the peak force that was in play to get the rotation occuring.

 

Now it's important to also note the radius of the shaft. It will be both numbers you need for specifying the torque and choosing the motor you'll need.

 

For example, suppose you measure 3 Newtons of force (that which the force gauge gives you). And the shaft radius is 5mm. In simple terms Newtons equates to Kg so the torque for the shaft load in this example would be written as:

 

   3Kg . 5mm

 

Now you need to normalise these numbers. First convert mm to cm or whatever metric the motor is specified in (typically cm): 5mm = 0.5 cm, so the numbers would be rewritten as:

 

  3Kg . 0.5cm

 

Now importantly the dot between these numbers doesn't mean "per" (as in Kg per cm). It means "cross product", however you don't need to worry about the details of what that means. Just how to get the numbers into the form required to choose the right motor. Basically you just multiply Kg by cm:

 

  3 x 0.5 = 1.5

 

So the torque becomes written as:

 

  1.5 Kg . cm

 

And there you have written down which motor you need.

 

C


Edited by Carl Looper, 19 November 2015 - 01:44 PM.

  • 0

#3 Simon Lucas

Simon Lucas
  • Basic Members
  • PipPipPip
  • 100 posts
  • Other
  • UK

Posted 24 November 2015 - 04:54 PM

Carl, thanks for the very precise explanation of how to do this. In the UK force guages seem to be hundreds of pounds, sterling. I can look out for a sub- £20 gauge but perhaps there is another way. You could tell me what the torgue was that you needed in your elmo, and I use that as a 'rule of thumb' when guessing what to buy :)

 

Steppers are in the range of £12-20 with a holding torque of 1Kg/cm. I'm not sure which torque measurement counts for this job but I would guess its the 'pull-out' torque. In the case of one I am looking at it has a 'pull-out' torque of between 300 and 500 g/cm.


Edited by Simon Lucas, 24 November 2015 - 05:07 PM.

  • 0

#4 Carl Looper

Carl Looper
  • Basic Members
  • PipPipPipPip
  • 1367 posts
  • Digital Image Technician
  • Melbourne, Australia

Posted 25 November 2015 - 04:57 AM

When I did the Elmo I didn't have the force gauge. I'd yet to learn that trick. As a result I got a stepper motor that didn't quite have enough torque. So I went and bought the biggest fattest stepper motor I could find. Which turned out to be so powerful I don't actually have anything for which it (rather than something smaller) would be needed!

 

I'll measure up the Elmo and give you a figure. I don't have the gauge here at the moment. But I'll get hold of it in the next few days.

 

C


  • 0

#5 Carl Looper

Carl Looper
  • Basic Members
  • PipPipPipPip
  • 1367 posts
  • Digital Image Technician
  • Melbourne, Australia

Posted 25 November 2015 - 05:11 AM

Re. step size. Doesn't really matter. Your motor shaft might move the film one frame for one or two rotations of the shaft. So if you had 200 steps per rotation that's heaps more steps than you would actually need. However you'll probably need to put in an additional switch of some sort to let you know when one frame has been moved. This is because although you can position the motor exactly, the belt attached to the motor and driving the film transport may not be so precise/consistent. You might find it takes 1.9 turns of the shaft to move one frame, but the next day, or a half an hour later later, its averaging 1.87 turns of the shaft to move one frame. It may be consistent or it may be inconsistent. You don't know.

 

So the idea is you create a feedback system whereby you simply stop the motor whenever a switch is triggered. How you implement that switch can vary. Could have a light sensor on the shutter blades. Or a Hall sensor on some shaft that's directly driven by the film transport. Or even a mechanical switch of some sort. When triggered you stop the motor, snap a frame, and then start the motor again until the switch is triggered again.

 

The stepper motor is just giving you the ability to stop and start transport willy nilly. The actual step size isn't really playing much of a role.

 

C


Edited by Carl Looper, 25 November 2015 - 05:18 AM.

  • 0

#6 Simon Lucas

Simon Lucas
  • Basic Members
  • PipPipPip
  • 100 posts
  • Other
  • UK

Posted 26 November 2015 - 07:21 PM

When I did the Elmo I didn't have the force gauge. I'd yet to learn that trick. As a result I got a stepper motor that didn't quite have enough torque. So I went and bought the biggest fattest stepper motor I could find. Which turned out to be so powerful I don't actually have anything for which it (rather than something smaller) would be needed!

 

I'll measure up the Elmo and give you a figure. I don't have the gauge here at the moment. But I'll get hold of it in the next few days.

 

C

Carl. Thank-you for your help - I'd appreciate it if you can find time. Did you find the Elmo OK for converting? And do you think it is possible to remove and enlarge the aperture in the gate?


  • 0

#7 Simon Lucas

Simon Lucas
  • Basic Members
  • PipPipPip
  • 100 posts
  • Other
  • UK

Posted 26 November 2015 - 07:26 PM

Re. step size. Doesn't really matter. Your motor shaft might move the film one frame for one or two rotations of the shaft. So if you had 200 steps per rotation that's heaps more steps than you would actually need. However you'll probably need to put in an additional switch of some sort to let you know when one frame has been moved. This is because although you can position the motor exactly, the belt attached to the motor and driving the film transport may not be so precise/consistent. You might find it takes 1.9 turns of the shaft to move one frame, but the next day, or a half an hour later later, its averaging 1.87 turns of the shaft to move one frame. It may be consistent or it may be inconsistent. You don't know.

 

So the idea is you create a feedback system whereby you simply stop the motor whenever a switch is triggered. How you implement that switch can vary. Could have a light sensor on the shutter blades. Or a Hall sensor on some shaft that's directly driven by the film transport. Or even a mechanical switch of some sort. When triggered you stop the motor, snap a frame, and then start the motor again until the switch is triggered again.

 

The stepper motor is just giving you the ability to stop and start transport willy nilly. The actual step size isn't really playing much of a role.

 

C

I never thought about using a feedback system – I assumed this was used by those who were using continuous motors. That sounds much more reliable and free from creeping error. That's really cleared up an area of mystery for me! Thank-you.


  • 0

#8 Simon Lucas

Simon Lucas
  • Basic Members
  • PipPipPip
  • 100 posts
  • Other
  • UK

Posted 12 December 2015 - 03:25 PM

Carl, I hope this reaches you before you make any measurements for me.

 

I have found and bought a stepper. It has Holding Torque 90N.cm, Detente Torque 3.5N.cm 2A/Phase. I think it will enough.

 

I have found a £5 Eumig 502 projector and taken out the motor. I am planning how to mount the motor, feedback unit and LED light.

 

I hope to get this built over the holiday period.

 

Thanks for you input.


  • 0

#9 Carl Looper

Carl Looper
  • Basic Members
  • PipPipPipPip
  • 1367 posts
  • Digital Image Technician
  • Melbourne, Australia

Posted 25 December 2015 - 03:10 PM

Hi Simon,

 

sounds good. My force gauge is still out on loan. Must get it back.

 

C


  • 0

#10 Simon Lucas

Simon Lucas
  • Basic Members
  • PipPipPip
  • 100 posts
  • Other
  • UK

Posted 31 December 2015 - 05:58 AM

Hi Simon,

 

sounds good. My force gauge is still out on loan. Must get it back.

 

C

 

There's a really bad Star Wars joke in there somewhere. Luckily, I can't think what it is.


Edited by Simon Lucas, 31 December 2015 - 06:04 AM.

  • 0

#11 Josh Gladstone

Josh Gladstone
  • Basic Members
  • PipPipPipPip
  • 309 posts
  • Editor
  • Hollywood

Posted 10 January 2016 - 03:24 AM

Yeah Simon, you basically want to come up with a way to know when one full revolution has occured, and you stop the motor at that point, either with an interrupt, or by just constantly polling whatever sensor you have.

 

The first way I did it was to hack apart a mouse, and use the mouseclick to trigger the image capture. This did work pretty well at first, but ultimately because there's a physical thing pressing against a physical part, it wears out and breaks. And because one roll of film is going to click that thing something like 3000+ times, it wore out pretty quickly. So ideally you want a way to detect when a frame is finished being pulled down and is ready for capture without physical contact.

 

Then solution I came up with was a photoresistor behind a hole behind the shutter, with an LED on the other side of the shutter. When the shutter is not between the LED and the photoresistor, it detects a lot of light, when the shutter passes between them, it detects less or no light. Every three shutter passes equals one frame, and so my program counts three shutter passes and then stops the motor.

 

I got this idea from looking at the Müller HM Data Framescanner, which was the initial inspiration to try and build a scanner in the first place. But there may be other better ways to detect the position of the stepper motor. A hall effect sensor - magnet combo, possibly? Or a rotary encoder on the motor itself? Lots of possibilities. I do plan to look into it more some day soon. 


Edited by Josh Gladstone, 10 January 2016 - 03:25 AM.

  • 0

#12 Simon Lucas

Simon Lucas
  • Basic Members
  • PipPipPip
  • 100 posts
  • Other
  • UK

Posted 10 January 2016 - 08:47 AM

Yeah Simon, you basically want to come up with a way to know when one full revolution has occured, and you stop the motor at that point, either with an interrupt, or by just constantly polling whatever sensor you have.

 

The first way I did it was to hack apart a mouse, and use the mouseclick to trigger the image capture. This did work pretty well at first, but ultimately because there's a physical thing pressing against a physical part, it wears out and breaks. And because one roll of film is going to click that thing something like 3000+ times, it wore out pretty quickly. So ideally you want a way to detect when a frame is finished being pulled down and is ready for capture without physical contact.

 

Then solution I came up with was a photoresistor behind a hole behind the shutter, with an LED on the other side of the shutter. When the shutter is not between the LED and the photoresistor, it detects a lot of light, when the shutter passes between them, it detects less or no light. Every three shutter passes equals one frame, and so my program counts three shutter passes and then stops the motor.

 

I got this idea from looking at the Müller HM Data Framescanner, which was the initial inspiration to try and build a scanner in the first place. But there may be other better ways to detect the position of the stepper motor. A hall effect sensor - magnet combo, possibly? Or a rotary encoder on the motor itself? Lots of possibilities. I do plan to look into it more some day soon. 

 

Josh - thanks for your input. I now have a basic mechanical system working - without the light and optics. A lot of it was inspired by your posts, along with some ideas from carl Hooper. I certainly got the motor spec from you and perhaps the idea of the photo-diode.

 

Here's the initial build. 

 

IMG_5178.jpg

 

Here's the one with the LED and photodiode that gives feedback to the Arduino.

 

IMG_5188.jpg

 

I'm now thinking about the light source. I was going to run an LED off the Arduino but I had no idea that LEDs could be so complex - with power ratings, current draw and heat dissipation important considerations. Also round LEDs and flat LEDs that affect the evenness of the light.  

 

I presume that with monochrome the key factors are brightness and evenness.

 

I have seen LEDs which have lenses and wonder whether this could contribute to an even light distribution?

 

I found these Luxeon LED assemblies, and I'm currently wondering whether they might be a good solution...

 

http://www.luxeonsta...-rebel-plus-led


  • 0



Technodolly

CineTape

Paralinx LLC

Tai Audio

Ritter Battery

The Slider

Willys Widgets

CineLab

Aerial Filmworks

Zylight

Abel Cine

Visual Products

Rig Wheels Passport

rebotnix Technologies

Glidecam

Broadcast Solutions Inc

Pro 8mm

Technodolly

rebotnix Technologies

Aerial Filmworks

Willys Widgets

CineLab

Paralinx LLC

Glidecam

CineTape

Rig Wheels Passport

Pro 8mm

Visual Products

Ritter Battery

Tai Audio

Zylight

The Slider

Abel Cine

Broadcast Solutions Inc