Step 9: Electronics

The roller switch is activated by the ramps built into the sprocket roller. The Arduino is used to activate the relay. The relay activates the camera shutter cable when the roller switch changes states. See the last photo in this step for a diagram of all wire connections.

  1. Mount the roller switch to a piece of acrylic or wood and use a 1.25" standoff to mount the switch arm to the gate platform. NOTE: the standoff height changes depending on the gauge of film your scanning. A higher standoff is used for 16mm than 35mm - depending on how accurate your legs are on the gate platform you may have to cut or extend yours.
  2. One a breadboard or a blank prototyping Arduino shield (pictured), wire the relay. In my case, I'm using jumper wires and connecting everything with alligator clips so it can be quickly disassembled if necessary.
  3. Splice your camera's shutter cable. You will likely find 2-3 wires inside. As a test, plug in the cable and turn your camera on. Find out which two wires, when crossed, trigger the shutter. If there is a third wire, it might be for triggering the auto-focus mechanism and should not be used. I bent it backwards and wrapped it in heat-shrink tubing. The two wires you will be using should be soldered to pins so that they can be plugged easily into the relay circuit.
  4. Connect the roller switch's connections to the circuit (see the drawing for a map of all connections).
  5. Plug in your 12VDC power supply to the circuit.
  6. Connect your camera to its cable and make sure all connections to the circuit and Arduino are complete, especially the connections to common ground. To test the circuit, plug in your Arduino to your computer (for power) and hit the roller switch. If the camera takes a picture then everything is working! If not, go back and check your wiring.
  7. Using the diagram provided, make your motor circuit and hook it up to power and Arduino. You can incorporate this circuit into the relay circuit or keep it separate. Either way works. This circuit controls the motor using PWM (pulse-width-modulation). It changes the speed of the motor to meet the frames-per-second variable set in the software. In some cases, you may want to control the motor manually. While this is a feature I plan to integrate, it is not currently available. To have full manual control, hook up your motor to an adjustable power supply according to its power requirements and adjust the voltage to change the speed.

You are now ready to operate the Kinograph. Operating Instructions coming very soon (as of 2/23/2014)

<p>Hi, </p><p>Awesome project! I noticed you were concerned about the shutter count on the 5d. The Magic lantern firmware has a feature called silent shooting. It essentially takes a picture from the live view keeping the shutter up the whole time. That might help extend the life of the camera beyond 200 minutes of film.</p>
Thanks for the tip. Do you have a link to the documentation on that? I'd like to see if the plane shutter is still used.
<p><a href="http://wiki.magiclantern.fm/userguide" rel="nofollow">http://wiki.magiclantern.fm/userguide</a></p><p>It is in the table of contents under -shoot -silent pictures.</p><p>In my own research, it seems to take stills from the video footage, meaning the highest resolution is around 1920 by 1080p. However, this software is also open source and the developers are very helpful, so you could more than likely ask them about any works in the future for that part of the program. </p>
<p>now magic-lantern suport hi res silent picture</p>
<p>Hi Again, would you happen to have obj or any other 3d file for super 8 sprockets? Many thanks!!! Richardson</p>
I'm working on those right now and will offer them as a Kickstarter reward (cheap) when the 8mm prototype is ready in a couple of months.
<p>I'm very much interested in Kinograph8. Where can we find more information? Thanks. </p>
<p>When will the Kickstarter reward be available? I'm currently working on one using a Raspberry Pi and am trying to figure out what I'll do for the sprocket. Also it might be worth it to put your 3D files on shapeways.com where people can order their own for those that don't have easy access to a 3D printer.</p>
<p>There hasn't been a kickstarter yet, but hopefully some time this year. I hope to send an update out to my mailing list (info@kinograph.cc) soon with more details. The 3D files are downloadable on the site (http://kinograph.cc) and you can easily upload that to shapeways if you'd like to get some made. Let me know if you do!</p>
<p>Would it be possible to post the rollers in .3dm format or similar? When I import the .obj files, they're not able to be edited to put in the Super8 sprockets.</p>
<p>I made a zip file for you with a 3dm file. It contains the final roller, the bump and sprocket parts I used to make that roller, and a blank roller for you to play with. I also included the python scripts I used to place the bumpers/sprockets on the roller. When you use it, place the blank roller at the center with its bottom on the plane. Link: kinograph.cc/files/rhino_roller_python.zip</p>
<p>Looking at it, I think the curve from the 16mm to the 8mm could be cut down (or removed) and then that might possibly give enough room to have the Super8 teeth on the opposite side from the 8mm and give each a lip under the other's teeth? I've no idea how to start editing the schematic in Rhino though.</p>
<p>Shapeways seems to be somewhat expensive at about $31-32 each. I found http://i.materialise.com/ and they can do them for a lot cheaper (around $14 for the sprocket and if you get 3 of the idle, those go to around $12). It'd be nice if somehow the Super8 sprocket could be added though.</p>
<p>I have created a forum for this project. It will allow users to share pictures ideas and upload files etc. I have started various forums in the past, some still active with more than 30,000 members. While I dont think this project will generate this many members I thing it should help refine the project and become a great place to share ideas. Please join in http://kinograph.freeforums.org/index.php</p>
<p>Thanks! To anyone looking, the official forum can be found here: http://forums.kinograph.cc/</p>
<p>Hey, I was just curious as to how you were able to focus the Raspberry camera at such a close distance? I am having trouble getting a clear image when the camera is so close to the film gate.</p>
<p>Hi @ksinger2. This version uses a DSLR with a macro lens. For the Raspberry Pi camera, you can try a couple of different things. 1st, try this: https://www.youtube.com/watch?feature=player_embedded&amp;v=FPv1sMk_fv4&amp;t=42, or you can buy cameras with lenses, like this one: http://www.mpja.com/Camera-Module-5M-Pixel-for-Raspberry-Pi-C_S-Mount-Lens/productinfo/31497%20MP/?gclid=Cj0KEQiAycCyBRDss-D2yIWd_tgBEiQAL-9Rku-lYe81oS8m_FzdEIeJTJtXDA_P9IYprCvQ6ecnHmwaApIi8P8HAQ that can be adjusted in a similar way (by tightening or loosening the lens in its mount).</p>
<p>Why not developping for HD moving camera. you can get the luma signal for synchro (from the yellow connector). I use that with a modified projector </p>
<p>Frederic,</p><p>I'd be interested in hearing more about this. If you're talking about adding a movie camera, this would require the frame rate of the video to be perfectly in-synch with the frame rate of the machine. This is difficult to do cheaply because it would involve fabrication of parts to create intermittent motion. Right now the film never stops moving, which is easier to manage. Is there a way to achieve the speed improvements of using a video camera without having to build an intermittent gate? OR - the question becomes how to build an intermittent gate that is reliable, cheap, and can be manufactured easily.</p>
<p>Hi I am about to embark on a similar project. I have some ideas that you might find useful (or not).</p><p>I read your comment about limited shutter life. (300,000) How about using an HD webcam and using the computer to capture the actual screen shots. As the film passes through an optical sensor/trigger it tells the computer to take a snap shot. On my machine each frame will temporarily stop for the shot. I will be using geared down stepper motors for the reel drives. There is software that can automate the process continuously. (Actionaz)<br>Also its possible to use a digital movie camera to stream continuosly to the computer and the same method of screen capturing can be used. This solves the whole frame rate issue.</p>
<p>That sure would speed things up with scanning. Tell us more Fred.</p>
<p>will you charge your device for a video camera.</p>
<p>using a video camera would require the frame rate of the video to be perfectly in-synch with the frame rate of the machine. This is difficult to do cheaply because it would involve fabrication of parts to create intermittent motion. Right now the film never stops moving, which is easier to manage.</p>
<p>You should make a 3d printer file of this. One part could be the back part, and the front part could have little bendy tabs. it could then fit toghether? Going to try to hack something togther trying to build this thing but going to use MDF so my design is going to be much cheaper haha.</p>
<p>This is a great project! Just a few thoughts:</p><p>You might want to check out the following page for some thoughts on the lighting and capture mechanisms. I'd be interested if the LED arrays and diffuser are something that could be worked in to your design.</p><p>It also has links to &quot;Fred's&quot; page, who has some pretty extensive post-processing scripts developed in AVISynth. I'm not sure if these functions would be useful or complementary to the software you're using.</p><p>I'd love to see your design able to support a range of capture devices - from cheap single LED light source and smartphone/Rasp PI camera all the way up to custom LED arrays (including IR) and machine vision cameras. </p><p>Of course the first and seemingly unsolved part that your project can accomplish is a reliable and cheap transport mechanism plus open source (free) capture and control software.</p><p>cine2digits.co.uk</p>
<p>This project is awesome! Btw, what do you think about using a pentax k1 (mirrorless interchangeable lens, </p><p>http://www.cnet.com/products/pentax-k-01/</p><p>) and a IR controller (</p><p>http://www.amazon.com/Wireless-Control-MagicFiber-Microfiber-Cleaning/dp/B0040RDT0M/) ? thanks!</p>
Yes, that could work. The roller switch is just activating a relay. The relay simultaneously turns on the LED and sends a pulse to the camera cable. If you want to switch out the camera cable for an IR transmitter, the rest of the circuit is unchanged. And yes, the Pentax should work great.
<p>If you were looking for capture devices I discovered a company called Ambarella that makes all kinds of video capture SoCs for all kinds of industries and at least one of them captures 4K video at 30fps and has both HDR and WDR tone mapping. I was thinking this might be a nice device to exploit for this project.</p><p>http://www.ambarella.com/uploads/docs/A9%20Product%20Brief%20120813-2.pdf</p>
This company looks really promising. I'll be contacting them this week. Thanks for the tip! Do you work with camera boards? My email = info@kinograph.cc
<p>On the project page you are using a DSLR. Could a modified Raspberry Pi camera module that has already been done for macro photography be used instead?</p><p> This would reduce the cost dramatically, and allow the unit to be entirely self contained by moving the capture device (camera) pretty close to the gate.</p><p>I was thinking about this like a VCR, or other similar device. Keeping the media close to the capture device.</p><p>I assume there are technical reason why this or similar solution wasn't chosen.</p>
It is a great idea, and one that I'm working on for the 8mm version of the Kinograph at the moment. There are problems with quality, speed, and the type of sensor that the small board cameras pose, however. Firstly, the dynamic range is pretty horrible. The shutter is a rolling shutter, which means you can't move the film very fast at all, especially when it's so close to the camera because you will get a &quot;jell-o&quot; or smeary effect. I think it could work well for small gauge film, though, and will be sending updates from the Kinograph email when I have some result. Send me a quick email there and I'll add you to the update list: info@kinograph.cc
<p>ofey/ trossi,</p><p>What is the unit measurement of the obj files pl.</p>
<p>I may have completely overlooked this, but are there any plans to support digitizing the sound tracks along with the frames? I have 16mm films that I would like to preserve, but would also like to eventually capture the sound. Not sure if there are any projects out there to convert the sound tracks.</p><p>This is a great project and I have been watching the progress for some time. I plan on using this setup with a Raspberry Pi and the Raspberry Pi Camera. Thank you for posting the 3D files for the rollers, I just printed the one with the sprockets and it turned out great. </p>
The sound is digitized using a piece of software developed at the University of South Carolina called AEO-Light. More info on that here: http://imi.cas.sc.edu/mirc-software-overview/<br><br>As soon as I get a spare afternoon I'll put together a user guide that walks you through the whole process.<br><br>I've been drafting plans for an RPi version and very much look forward to seeing how it works for you. Thanks for following and good luck with your build.<br><br>-Matt
This is an awesome looking unit. I would recommend the use of the word telecine in your build description. I've been working on my own machine for years and there is a substantial community out there who would be interested. I just can't say as I've ever he heard the term Kinograph use in the community before. Awesome build though. I noticed much of the cost is in the frame. I assume you built it out of aluminum was for transporting it mostly? Of course it looks awesome.
Yes, you're right. The frame is very expensive and truthfully does not need to be made out of extruded aluminum. My choice for that material was based on flexibility since, as I was building it, my designs kept changing. The slots in the aluminum make it easy to for adjusting the position of parts without having to drill. The cost could come down significantly by building the frame with more common hardware found at hardware stores.

About This Instructable




Bio: Blogging at: http://reddesert.tumblr.com/
More by mepler:Kinograph v0.1 - DIY Film Scanner/Telecine - Machine Assembly RaspTI: Convert a Vintage Computer (TI-99/4A) into a RaspPi Workstation - Part 1 - Keyboard Make Silicone Molds for your 3D printed object 
Add instructable to: