DIY Crop Sensor Slide Duplicator

From Dallas Makerspace
Jump to: navigation, search

Adapting a Vintage Slide Duplicator for use on a Crop Sensor DSLR

Steve Rainwater

When faced with the task of scanning large numbers of 35mm slides as part of a project to digitize my family's photographic history, I began pondering ways of speeding up the process. High resolution scans of 35mm slides on either flatbed scanners or dedicated film scanners is a very time consuming process. As a collector of vintage camera gear, I was aware that slide duplicators existed for 35mm film cameras. These devices were mounted to the camera either on the end of a lens or in place of a lens with the goal of allowing a 1:1 copy of a 35mm film frame. With the proper adapters such duplicators could theoretically be used on a full-frame sensor DSLR such as the Canon 5D.

My budget only allows for the use of less expensive crop-sensor cameras such my current Canon 40D. The 40D has an APS-C sensor, which is about 22.2mm x 14.8mm in size. A full frame is 36mm wide. This difference in size is used to compute the "crop factor (e.g. 36mm / 22.2mm = 1.621621...), so in effect the smaller sensor acts like a 1.6x magnification factor, making the focal length of lenses seem larger than they really are. A standard 50mm lens acts like an 80mm telephoto lens on a 1.6x crop sensor camera. So a vintage slide duplicator would be affected similarly and the result would be a cropped section of the center of the slide you wanted to duplicate.

It seemed like there had to be some sort of do-it-yourself solution to this problem and a little searching turned up cases of at least partial success in adapting duplicators for use on crop frame cameras. One example is the adaptation of an Acura Slide Duplicator to Pentax SF10 DSLR with a crop factor of 1.5x, described in a post on the Manual Focus Lens forum. The approach taken was to use a lens-mounted slide duplicator combined a shorter focal length lens and additional adapters to keep things spaced appropriately. I decided to start with the same approach and will document my experiments here for others to learn from.

Control Data

For testing purposes a representative slide was selected and scanned on both an Epson Perfection V500 photo flatbed scanner and on a high-end Nikon film scanner. The same slide will be used to test each rev of the slide duplicator. Attributes to be compared including distortions introduced by the lens, as well as color and sharpness of the resulting image.

Revision One

<flickr>6044014305|right|frame|DIY Slide Duplicator Rev. 1 <a href="http://www.flickr.com/photos/steevithak/sets/72157627435898510/">[more photos on flickr]</a></flickr>

Components used

  • Generic EOS mount 16.25mm extension tube
  • Canon EOS to Olympus OM mount adapter
  • Vivitar 28mm f2.8 Olympus OM mount lens with 49mm filter size
  • 49mm to 52mm filter ring step-up adapter (approximately 4.5mm depth)
  • 52mm to Series VII filter ring step-up adapter (approximately 6.6mm depth)
  • Series VII to Series VI filter ring step-down adapter (no additional depth)
  • Series VI retainer ring (approximately 8mm depth)
  • Accura Variable Magnification Slide Duplicator

Notes: The Accura Variable Magnification (VM) slide duplicator was designed to work with 50mm to 58mm lenses by means of a sliding extension tube with a set screw. In theory, this translates to a range of 32mm to 36mm lenses on a crop sensor camera. My goal was to use parts on hand if possible, so I started with the 28mm lens because it was available. Beside the focal length being a bit too short, there was a good chance of distortion from the wide angle lens. So consider this rev of the device a proof-of-concept.

Results: Mixed. The smallest extension tube I had available was 16.25mm. It claims to be 6mm but the front and rear mounts are normally not included in the tube's stated measurement. The extension tube is necessary in an arrangement like this to bring the focus range of the lens down to the distance of the 35mm slide, in this case about 2 inches from the front of the lens. Unfortunately, short focal length lenses are more dramatically affected by extension tubes than longer lenses. The shortest extension I had was too long, forcing the focus range to a point where infinity was about 1/2 an inch from the lens. Aside from the focus issue, however, the results were promising. The 28mm focal length is equivalent to a 45mm lens and, as expected, the field of view was slightly wider than the full 35mm slide. Fixing the focus problem would require switching from a commonly available extension tube to a more expensive extension ring designed for use with wide angle lenses. Surprisingly, it proved much less expensive to pick up 35mm lens than a 4mm-6mm extension ring. So, while not usable, the first rev of the device did prove that the basic concept could work.

Revision Two

<flickr>6044014307|right|frame|DIY Slide Duplicator Rev. 2 <a href="http://www.flickr.com/photos/steevithak/sets/72157627435898510/">[more photos on flickr]</a></flickr>

Components used

  • Generic EOS mount 16.25mm extension tube
  • Canon EOS to Nikon mount adapter
  • Soligor Wide-Auto 35mm f2.8 Nikon mount lens with 49mm filter size
  • 49mm to 52mm filter ring step-up adapter (approximately 4.5mm depth)
  • 52mm to Series VII filter ring step-up adapter (approximately 6.6mm depth)
  • Series VII to Series VI filter ring step-down adapter (no additional depth)
  • Series VI retainer ring (approximately 8mm depth)
  • Accura Variable Magnification Slide Duplicator

Notes: The big change in Rev. 2 is the use of a 35mm lens instead of a 28mm. I obtained a like-new Soligor 35mm lens on eBay for $9 (plus it came with 10 assorted 49mm filter for my junk box).

Results: Success! With the 35mm, the extension tube worked perfectly and brought the focus range down to an ideal range approximately in the center of the Accura's magnification range. There appears to be no visible optical distortion from the wide angle lens. My initial test slide showed slight but noticeable vignetting on the right side. The vignetting may be due to off-center placement of the test slide.

Revision Two Samples

<flickr>6078290301</flickr> <flickr>6078831688</flickr>
Sample shot taken in May 1981 with a Canon A1 and Canon 50mm f/1.4 lens. Scanned on Epson Perfection V500 Photo scanner. Sample shot digitized with a Canon 40D and rev 2 slide duplicator. Several flaws are evident including a color shift, a focus or resolution issue, and keystoning. There's also a problem with the duplicator's slide mount interfering with the left and right edges of the image.

References