The Scanner Camera

The Idea

This camera is the result of my longtime desire to spend less than 50,ooo dollars on a large format digital camera. In the end I spent about forty dollars in addition to a lot of time and parts that I already had. I got the idea when scanning LF negatives in college and wishing I had a digital LF. Three years later I finally had the time and inclination to try to combine a scanner and a LF camera to try to make something that has the focal control of a LF and the ease of use that comes with a digital image file.

Of course I had a romantic mental image of what would be, how easy it would be to produce, and the amazing quality and durability of the whole setup, but in the end my camera looks less like a sexy digital hasselblad and more like a ill-devised medical device.

A few searches in google and I had found the websites of a few successful DIY LF to Digital conversions. Some had gone crazy spending thousands, others were obviously no amature hackers, and some had less than desireable results, but I got the general idea. The concept behind a camera is that a series of ground glass elements form a lens which magnifies and focuses light onto a surface where the image is duplicated. Leave out glass elements from the previous description and you get a camera obscura. Add in a chemically reactive layer or a digital sensor to your image duplicating surface and you have the  modern camera.

The close interdependence of spacing, lens characteristics, and image recording device all factor in to the engineering of a camera and changing one element can affect the effectiveness of them all. Thus to sucessfully change a LF film camera to digital  I modified a scanner to accept more light and be more light tight, and I modified the camera to set the scanner as close as possible to the existing film plane to avoid cropping and focusing problems. One website creator that I had found in my searches had built his scanner camera project around a Canon Canoscan LiDE 20 and had gotten fairly good results. his website can be found here: http://golembewski.awardspace.com/cameras/distortion/index.html. Coincidentally I found the exact same scanner at Next Step Computing for $15 about a week later. The online guide linked to above shows how to remove excess material from the scanner to allow it to absorb more light, as well as to take reflection to a minimum. The scanner is easily disassembled, a good thing since I couldn’t find a guide on how to do that. Finally a light proofing was applied in the form of gorilla tape and spray painted cardboard to make sure the camera got only the image it was supposed to.

To have the least light drop off, as well as to maximize focal range and ensure ease of mounting I removed everything I could from the back of the field camera body. The bare aluminum was all that was left when I was done and with a bit of taping I was sure that light wasn’t coming in where it wasn’t supposed to.

Finally the two pieces were mounted together and viola; a camera is born!

I’ll leave the details of creation for the body of the post. If there is anything that you don’t find here please check out the link above, or in the resource list at the bottom. If you can’t find it there feel free to email me at eschafermeyer<at>gmail.com. Thanks!

Materials

1 Burke & James 4×5 Field Camera

3 Canon Canoscan LiDE 30 Scanners

1 Roll Gorilla Tape

1 Cardboard Box

1 Roll Electrical Tape

5 Linear feet of hardwood flooring

1 unit graphite lubricant

Assorted Fasteners

Tools

Utility Knife

Dremel w/ fixins

Drill/Impact Driver

Screwdrivers

Pliers

Funny Hats

Tequila

The camera I had. The first scanner I got from a local recyclers called Next Step Computing. I paid $15 for the Canon Canoscan LiDE 30. Max resolution 1200 DPI, CIS Sensor, USB Powered. All around a small and simple machine. I broke that one fairly rapidly and found myself traveling to stumptown where a craigslist add found me two more for $20. After that the project was on. Here is a fairly (un)detailed description of the process, my resources, and the results.

The Camera First, the camera. This is a Burke & James 4×5 field camera; the type used by press men back in the day when it was acceptable to refer to them as “pressmen.” This was a hand-me-down and despite it’s steampunk value to me I decided to sacrifice it in the name of the scanner camera project. In the second photo you can see that I removed the clip from the back of the camera which held into place the film carriers and housed the ground glass. this was held on by screws and a bit of glue. Also, I years ago had removed a periscope-like viewfinder from the left hand side (facing) of the camera. I did not remove the solenoid from the front of the camera which fires the shutter with the charge of a 9v battery. Not sure how that will help as I will be leaving the shudder open so that the scanner can travel its entire path.

The Scanner

I will leave most of the details of how to modify your scanner to the webpage from which I learned it: http://golembewski.awardspace.com/cameras/distortion/index.html. Here is a little on that subject however:

This is what the scanner looks like. On the right and under the cover we have three modifiable buttons which allow you to scan and save in different ways. On the left side there is a USB port. This scanner is USB powered which is Awesome!| b/c it allows me to take the camera into the field and shoot with only it and my laptop. I am using a Mac with VueScan software.

The inside.

Here you can see that i have peeled up on each of the gray plastic strips holding the glass in the scanner. I started peeling from the side of the scanner nearest my hands, peeing slowly away. Then I pushed the glass away from me and pulled it from under the plastic lip. Thus providing access to the interior of the scanner.

First I pulled the scanner assembly away from its berth under the plastic lip which houses the three black buttons. Pulling the assembly causes he threaded chord to move along the gearing and subsequently move the motor on the underside of the assembly. Doing this makes a bad noise but, if done slowly, seems to not damage the motor or the gears. I then removed the two small black pieces of  plastic which can be seen above from the black plastic bar which houses the LED assembly and CIS sensor(s?) (also above). These two small pieces of plastic are rotationally symmetrical (at 180deg) and are identical. They each have a small post at either end which lodges in the black plastic of the CIS housing. These two pieces must not be lost as the tension between them and the spring below them is what keeps the CIS sensor running on a parallel plane to the glass above it as the scanner runs. I ran my camera last night without one of the slider pieces in place and it was obvious immediately that the image was degraded on that side of the camera. Next I removed the black plastic bar which houses the CIS and LED assemblies from its mount on the aluminum mounting plate. This is achieved by bending back the right (in the above image) eye loop which holds a peg extending from the CIS housing, and removing a small ribbon from its connector on the underside of the black plastic bar. You can see the disconnected ribbon in the image above on the right side of the green PCB.

Modifying the CIS Housing

The goal of modifying the CIS housing is to reduce glare, destroy the LEDs and the LED assembly, and most importantly to increase the amount of light which reaches the CIS by removing the pinhole array and grinding the CIS housing down to expose the sensor. For more on this read further on the above referenced site.

This is the removed CIS sensor housing with the sensor and LED assemblies. In this photo i have carefully removed the clear plastic bar which distributes the LED’s lights across the object being scanned. The Plexiglas bar can be removed carefully by scoring the glue which holds it in with a utility knife until the bond is broken, the bar should lift out at this point. On the right side of the CIS housing the small white bar is the enclosure for the LEDs. At this point you can also remove the black plastic strip which is next to the depression that the Plexiglas bar sat inside of. The black strip is actually a linear array of pinholes designed to focus light from the scanned object onto the CIS sensor. Since you are going to be attaching a lens assembly in front of the sensor this pinhole array is superfluous and it reduces the amount of light which reaches the CIS sensor.

Now that the plexiglass bar and pinhole array are removed flip the housing over and start to cut away at the plastic which is holding in the PCB board.

In this image you can see that I am grinding down the contacts which hold the PCB in place. This should be done carefully and thoroughly as this is the first (and so far only) time that I broke a scanner. Specifically I pulled on the PCB thinking it was fully removed and I snapped the board, super lame.

Here you can see that I successfully removed the PCB and from the enclosure. Before replacing the PCB into the enclosure I snapped the while LED enclosure off by bending it ever so gently back and forth until the wire connectors severed. Those with more soldiering skills may wish to simply un-soldier the connection and not risk the posibility of breakage.

Here you can see that I have carefully, and with much help from Michael and a vacuum cleaner, ground down the walls along the side of the slit through which the CIS sensor gains light. Then the slit itself can be widened with a utility blade and sanding paper. After you grind away enough of the housing clean it before reinserting the CIS sensor assembly (sans LED). The sensor can be taped in place with a couple of short pieces of electrical tape.

With a bit of taping and some matte black spray paint the silver motor assembly can be made black as shown in the picture below. This is best done after first removing the motor assembly from the scanner shell. I don’t yet have any photos of how to achieve this removal but it can be done by first unwrapping the threaded chord from the white plastic gears under the camera and then lifting the motor assembly from the shell. The aluminum can be removed from the gears, electric motor and PCB by a few screws and some clips, just go slow, you will figure it out.

Here is the reassembled (and newly blackened) motor assembly with the CIS sensor housing replaced, all remounted into the scanner shell. Congrats, that was the hard part.

Mounting the Scanner to the Camera

Now that the scanner is properly modified it must be made ready to mount to the camera.

Here you see the shield that Michael and I made to stop light from coming into the scanner from all but the camera itself (and hopefully just the lens at that). I attached the shield with Gorilla tape which is thicker than duct tape and therefore hopefully more light-tight. Next I used some hardwood flooring scraps to create rails along which the scanner can slide down to mount on the camera onto which I mounted grooves. For all you aspiring carpenters out there, remember: Measure twice cut once. Then cut again. Then…. whatever, it worked out eventually.

And the back side:

Then I mounted it to the camera:

And now the finished product:

Glorious. I know.

So, I finished the build about a week ago and since then I have been able to go out a couple of times and shoot a little. I will post a follow up blog shortly about what I have learned and what I have gotten image wise. Thanks for reading the build blog. I hope you found it helpful. Please feel free to contact me through this blog if you have any questions.

-Erich

Advertisement