Announcement

Collapse
No announcement yet.

Homemade Telecine

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • #61
    Removed the Wolverine power supply from the rig and added my own 12V dc adapter and a small hardwired regulator board that interfaces with the Hawkeye board. The 12V provided to the Hawkeye 3.3V regulator is too high and causing it to heat up. So decided to go with a new regulator design.
    Click image for larger version

Name:	power.jpg
Views:	405
Size:	71.1 KB
ID:	45124

    The regulator provides 5V for the Hawkeye board (will become Hawkeye2) and the 12V unregulated can be used for the stepper controller.

    Comment


    • #62
      I'll just wait for the parts list when you're done on the electrical!

      I'm still fiddling with the pinch and motor rollers. I assume you're keeping the light sensor that's attached or part of the sprocket?

      I thought about putting the motor mount and pinch roller support together. I need that file (I might already have it) for the sensor wheel, unless you think that might change.

      Click image for larger version

Name:	PINCH2.jpg
Views:	367
Size:	60.5 KB
ID:	45273

      Comment


      • #63
        There will be two schematics. One for the V12 Hawkeye version and another for Hawkeye2.
        So, you plan to do the motor mount and the pinch arm together as one part or a few parts connected together. I assume something similar to kinograph.
        Currently, I am not using the cogwheel and the optical sensor with the pinch roller. Don't think it will give us much improvement. Most of the shake is coming from the stepper accuracy in my opinion. Going with the 256 step driver could provide some improvement but I do not think that is necessary since the deshaker does a pretty good job.

        Comment


        • #64
          Tested the new stepper controller and new power supply and it seems to work ok. Will have to check the frame accuracy tomorrow. There are a few settings on the controller that could impact accuracy so will play with that.
          https://photos.app.goo.gl/e4yx5Twa9sygY21m9
          If it works OK will clean up the test rig and then make something close to kinograph but no raspi and raspi cam. Use MSP controller and good quality cam and lens.

          Comment


          • #65
            This motor (or similar) can be used for the takeup. It is 12V. A clutch similar to Wolverine would have to be designed to go with it.
            https://www.amazon.com/Bringsmart-JG...dp/B078JHVJYL/

            Comment


            • #66
              I'm hoping to be able to re-use the stepper and takeup motors as well as the reel posts from my Winait parts (same as the 720p Wolverine). It would be nice if I can leave them mounted to the original face plate.

              I spent too much time adding the pinch roller to the motor support I have. I really need to create a layout for all the parts first.

              Trying to build an assembly in Fusion360, I realized I had to go back to school (Youtube)! Drawing single parts to print has been easy.

              The /Alexamder/rpitelecine project documentation does not go thru all the details of the mechanical parts and assembly. I can work something out from the pictures. Creating the documetation must be a lot of work. I was spoiled by the documentation for the Hawkeye!

              Judging by the size of the 256 step driver's box, there should be plenty of control inside?

              Comment


              • #67
                The Wolverine parts should work ok but it is a question of mechanics.
                I am trying not to use the Wolverine parts so that people that do not own a Wolverine could make this but running into issues with the new stepper controller. It seems to have some issues with accuracy. I get close to a sync and then it goes off either way.
                Will have to check the mechanics again to make sure everything is nice and solid.

                OK solved the issue. The stepper roller had some play in it. Also the spring tension was not enough causing some slippage.

                Here is a quick way pf checking the spring tension.
                Add another spring and check if the sync changes. If it does then you the spring tension is not enough. Add one more spring or use a stronger spring.
                Here is a test clip of the capture. It jumps around but that is normal, the deshaker takes care of that. There is almost zero long term drift.
                https://photos.app.goo.gl/5euoJri1wnPqK4KPA
                Cleaned up the unit a bit.
                https://photos.app.goo.gl/TmhAJJZCN3LpurLX7
                Last edited by Stan Jelavic; October 29, 2021, 03:27 PM.

                Comment


                • #68
                  I thought the same thing about this process, it should should have wide appeal.

                  No need for a Wolverne, and it looks pretty simple to make mechanically. I have some rewind arms from a broken editor/viewer I could use or a pair I printed that use 1/4" bolts for film reels.

                  Dare I say any camera can be used in that wide open space above your film gate.

                  I was looking at the 3D parts for the rpitelecene again and refocused on the gate. I'm going to try it first before tackling the S8 camera gate. It's not what I initially thought. Many homemade gates don't really hold the film flat. This one applies tension to the edges of the film from the sides. This is the kind of gate that could also adapt to baddly split 8mm film.

                  Interesting that though he included two 3D gate versions in the project, the picture is different. At least the bolts should have the nuts below, not on top.

                  Click image for larger version

Name:	cam-mount-3-assembled.jpg
Views:	326
Size:	42.5 KB
ID:	45763

                  Comment


                  • #69
                    Yes the idea is to have a simple unit that can be easily adapted to any camera. With some more work it could be made with interchangeable parts to work with any size film 8mm,S8 and even 16mm. For 16mm the gate would have to be swapped out and the reel mounts would change. I believe the pinch rollers could stay the same as long as they are long enough. The MSP can run different film pitch controlled by a switch.
                    The picture that you included does match the 3D design but I do not like it. The springs are designed to make the width flexible but there is no pressure on the film sides which could cause buckling of the film and focus issues.
                    In my opinion this below is a better design.
                    https://www.shapeways.com/product/9X...more-from-shop
                    This would still need some spring tension to be added.




                    Comment


                    • #70
                      Funny you should mention film buckling and focus issues! That's what I found in testing, but in the most extreme case.

                      8mm film width can range from 7.938-8.014mm (not including badly split R8mm).

                      Click image for larger version  Name:	super8smptespecs.jpg Views:	0 Size:	97.2 KB ID:	45989

                      The RPItlecine Gate3 has 7.90mm width spacing for film.

                      Click image for larger version  Name:	GATE3_RPItelecine.jpg Views:	0 Size:	20.9 KB ID:	45990

                      I made a handful of versions of this. With the original I didn't have any bolts long enough to add springs. I found by adjusting the bolt tightness film tension had a lot of range. Later, I saw the film would deform when the bolts were tight.

                      It's not likely that the gate would be tightened enough to deform the image like the extreme example shown below.


                      Reasonable tension Click image for larger version  Name:	image_10417.jpg Views:	3 Size:	122.0 KB ID:	45992 Bolts fully tightened Click image for larger version  Name:	Tight bolts.jpg Views:	0 Size:	101.2 KB ID:	45991

                      Splices do require more force to pass thru the gate. I don't know at this point if that will be a problem.

                      In my opinion the springs serve as a way to add or remove the film without loosening the bolts. The size and strength of the springs determine how easily the gate opens while keeping tension on the film.

                      My mod to this gate opens the width up to 8mm and the film cannot buckle, but has no adjustment for tension. The film is not lose and splices pass thru easier. I don't expect any recoil (with the pinch roller) like when the claw drops and the frame position jumps.

                      I have a working version so now I think I'll make a gate like Gavin's at Shapeway (link in #69).
                      Click image for larger version  Name:	IMG_20211101_130806744.jpg Views:	0 Size:	74.2 KB ID:	45994

                      Comment


                      • #71
                        Good work David. Here is a concept gate that I came up with. Still needs some more work.
                        Click image for larger version

Name:	film_gate2.jpg
Views:	277
Size:	74.9 KB
ID:	46036

                        Comment


                        • #72
                          Refined the gate design a bit and added the spring and bolt. The spring do not have to be 3D printed but are added for better understanding how this goes together. Click image for larger version

Name:	film_gate_concept.jpg
Views:	250
Size:	85.4 KB
ID:	46256

                          Comment


                          • #73
                            Not done with the film gate, but it's good enough to put aside and move on to making some film guides/pulleys. I have a 3' piece of 80/20 track that I bought 5 years ago as a mounting base for this same kind of project.

                            I once though that a horizontal mount was best. I going to stay with the vertical since I have a copy stand for my camera and these parts could sit below it on the table.

                            Lot to do, like adjustable mounting tabs above.

                            Click image for larger version

Name:	DSC07892.jpg
Views:	263
Size:	114.0 KB
ID:	46262 Click image for larger version

Name:	DSC07893.jpg
Views:	264
Size:	93.3 KB
ID:	46263
                            Click image for larger version

Name:	DSC07901.jpg
Views:	259
Size:	146.5 KB
ID:	46258 Click image for larger version

Name:	DSC07894.jpg
Views:	257
Size:	97.8 KB
ID:	46260
                            Attached Files

                            Comment


                            • #74
                              Good work. Yes you will need adjustable tabs, some type of spring action on the pressure plate, Will need the ridges in order not to scratch the film, lock channel on the pressure plate to prevent it from sliding lengthwise. Your picture with the mask inserted went out of focus. Did the pressure plate move when you inserted the mask.

                              Comment


                              • #75
                                The picture is out of focus, didn't notice until later. I should have posted the 3d pictures. The printed parts are a bit small and black. No mounting tabs yet as I don't know what it will mount to. I think it should sit on a small piece of wood with hole for an led as you have done. I've been using a small battery led light panel for light. It works well, but it's temporary.

                                Click image for larger version  Name:	Screenshot 2021-11-06 141249.jpg Views:	0 Size:	52.8 KB ID:	46309
                                This morning I started to revised it again!There is no need to have the mask fit in the bottom dovetail. I can move the mask up just under the film in a squared channel. The two halves screwed or glued together will hold it in place. The two halves is a printing issue. After printing so many versions, I chose to avoid supporting structures.Cleanup is tedious and can damages tiny surfaces.

                                I just now saw the spring you added to the pressure plate. I like that.

                                Comment

                                Working...
                                X