Jump to content

It's Time for a Real DIY MP Film Scanner


Anthony Schilling

Recommended Posts

Im really going to miss 100D in S8 and 16mm! But I'm happy the format is still going and getting V3 50D in S8. I have shot V2 50D in S8 and it's some of the best looking S8 i have seen. The V3 200T is incredible too. I would love to shoot it as much as I would shoot reversal, but there's no way I can shoot more than 8 rolls a year with Telecine costs the way they are. In this day and age, i don't think a consumer grade home scanner that loads film reels into a PC would be that hard to make. They had them for still film, why not amature MP? Amature video and music production has boomed thanks to new technology that bypasses expensive studios. S8 and 16mm could have a great new life if someone could fill the missing link! I think we should start a petition to pitch. It would also come in handy for mass amounts of old movie film sitting in most closets and atticks. That service Kodak tried to start was a flop from the start. Just give us a freakin device already!

Link to comment
Share on other sites

Actually, yes, Dragon would be an ideal building block for this sort of thing.

 

I do a lot of motion control and stop-motion work, and I've bee around Dragon a lot.

 

i actually looked at an application like this about a year ago.

 

It's a good startign point for this kind of stuff, since it handles allt he camera overhead plus has a reasonable motion control capability that lensd itself well to step-and-repeat functions.

 

You'll still have to make an "optical bench" of sorts, with an appropriate, reasonably well-registered movement and light source with long-term consistancy, but if you can hook a step motor up to it, then you won't have to re-invent the wheel as far as coordinating motor and camera and storing off the frames goes.

 

Dragon is inexpensive - somewhere around $300 IIRC. You can roll your own motor electronics, but if you don't want to do that, Dragon sells a unit called an Iota controller for $750 that can do all the motor driving (The Iota controller was originally designed to drive an I/O slider and focus motor, but it can drive any two small step motors) .

 

Two caveats - first, you'll probably want to arrange the gearing so that an even number of motor turns advances one frame (dragon step motor axes don't like fractional steps in the current software rev) and second, you'll probably have to work in blocks of a few hunderd frames at a time - because it's juggling allt he frames in memory, the program can bog down on really long animations.

Link to comment
Share on other sites

Actually, yes, Dragon would be an ideal building block for this sort of thing.

 

I do a lot of motion control and stop-motion work, and I've bee around Dragon a lot.

 

i actually looked at an application like this about a year ago.

 

It's a good startign point for this kind of stuff, since it handles allt he camera overhead plus has a reasonable motion control capability that lensd itself well to step-and-repeat functions.

 

You'll still have to make an "optical bench" of sorts, with an appropriate, reasonably well-registered movement and light source with long-term consistancy, but if you can hook a step motor up to it, then you won't have to re-invent the wheel as far as coordinating motor and camera and storing off the frames goes.

 

Dragon is inexpensive - somewhere around $300 IIRC. You can roll your own motor electronics, but if you don't want to do that, Dragon sells a unit called an Iota controller for $750 that can do all the motor driving (The Iota controller was originally designed to drive an I/O slider and focus motor, but it can drive any two small step motors) .

 

Two caveats - first, you'll probably want to arrange the gearing so that an even number of motor turns advances one frame (dragon step motor axes don't like fractional steps in the current software rev) and second, you'll probably have to work in blocks of a few hunderd frames at a time - because it's juggling allt he frames in memory, the program can bog down on really long animations.

 

I've just been playing around with a Canon DSLR, an old micro-Nikkor, a lightbox and some 35mm (stills) film, image reproduction is pretty good. I might have to try and rig something together, shame I missed buying a couple of back projection mitchell movements for peanuts a year or so ago. I have a bunch of cheap Chinese 15mm rod clamps and stainless tube, that would probably make an adequate optical bench.

 

I also have a background (if you can call it that) in motion control for stop motion, and have recently started playing with dragon, here's some kit I've built http://www.flickr.com/photos/mohansandhu/

Link to comment
Share on other sites

>> I've just been playing around with a Canon DSLR, an old micro-Nikkor, a lightbox and some 35mm (stills) film, image reproduction is pretty good.

 

You know what would work really well for this wouldn't be a micro Nikkor, it would be an old c-mount 25mm macro-Switar mounted backwards.

 

I played with this a few years ago for some micro work with a custom bellows system, using reverse-mounted 16mm lenses.

 

They don't have a big image circle, but if you're on a small subject they have phenomenal image quality over the area they do cover, and they're physically small enough to light around.

Link to comment
Share on other sites

  • Premium Member

Would be interesting if the actual film transfer houses could set up small rooms that have the basics for low cost film scanning and simply charge a rate of 25 bucks an hour. However, the downside could be that if it takes four to six times as long as a conventional transfer, it may mitigate the cost effectiveness to some degree.

Link to comment
Share on other sites

  • Premium Member

My feeling about this is that it's like a lot of DIY.

 

Building cables is a common pastime among camera people who can solder, but the idea expands to all kinds of stuff that people knock together in the back bedroom - fluorescent lights, even dollies, and now CMOS sensors are cheap, film scanners.

 

The problem is that it tends to work out OK if you do it for yourself. Trying to make a business out of it can be very difficult, and that limits the amount of engineering you can do.

Link to comment
Share on other sites

  • 2 weeks later...

this thing just appeared on Filmshooting.com. The only thing that interests me about it is the fact that it's sprocket-less, other than that the small reel capacity, fact it spits out jpegs and won't in it's current state scan negatives limits it's usefulness to me.

 

Also, if you're going to design a sprocket-less machine why not make it compatible with both super8 and 16mm?

Link to comment
Share on other sites

  • 2 weeks later...

I was thinking about cannibalizing a broken super 8 camera and rigging it up with some kind of hand-turn knob to advance the film. Then just use a digital camera to snap a photo of each frame. It might seems tedious, but it's way easier than scanning on a flatbed and cropping each frame by hand... which is what I've been doing.

Link to comment
Share on other sites

  • 1 month later...

There is a lot of interest in this topic these days and so it probably going to happen. It probably will be moviestuff. that retro scanner is V1.0, just wait until subsequent updates address many of the needs stated on this thread,(negative scanning, multi format, uncompressed output,). Others manufacturers are currently offering machines that seem to be geared toward labs and post houses, so a bit out of reach for the DIYer, which I include myself. The retro scanner with some upgrades again, might fill the void. I'd like to see at least two flash scanning and an IR flash. Optical panning and zooming would be great as well as anamorphic correction. The big issue in terms of cost is that Nikon, Canon, Minolta etc... could count on tens of thousands of units being sold. They also had all the infrastructure in place for R&D and manufacture. Also, a scanner that can do a multi flash scan and IR pass in close to real time, is more expensive to make than one that take several minutes per frame. I believe that a sub five thousand dollar scanner that can do all this is possible

Link to comment
Share on other sites

  • 2 weeks later...

I have built my own unit and it works pretty well. I would like to improve it to a pin registered system in the future.

 

Basically I used a Point Grey Grashopper Express 2.8Megapixel machine vision camera which can do 24fps at 1080P resolution.

I use Frank Vine's LED lighting system that gives high intensity white balances flashes @ 24fps, and a simple ($10) hall effect circuit to know when the frame is in the gate ready to image.

 

A reversed enlarger lens forms the imaging system and projects directly onto the chip of the camera.

 

This allows a full speed transfer with a high bit depth in real time. All up cost was about $3000, the camera being the expensive bit @ $2000.

Link to comment
Share on other sites

  • 2 weeks later...
  • 3 weeks later...

I would beg anyone who takes on this sort of project to release detailed designs under a creative commons license, ala Reprap 3D printers. Including directions for construction and technical drawings.

 

I suspect the proliferation of this kind of machine would open up a few more art projects to shooting film.

 

I wonder if 80/20 extrusion would be a good foundation for mounting the components?

Link to comment
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
×
×
  • Create New...