Announcement

Collapse
No announcement yet.

Wolverine MM Pro problem with Super-8 please help

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

  • #31
    Thanks for the inputs. It is just a concept or a discussion piece without serious intentions of implementing it. It is similar to some already existing designs but without the perforation detector. And you are right David, film shrinkage could impact its performance, although I believe it would just shift the frame that could be compensated for by using a camera micro adjuster. The takeup would have to apply a continuous tension so that the film sits snug around the capstan. Possibly something similar to Wolverine with a spring clutch.
    And yes an encoder can be implemented instead of the micro switch. The capstan can have a flange with the indentations that match the sprockets. An optical detector (similar part used in Wolverine) then could be used to indicate the beginning of the frame. We know that this is pretty accurate.

    Comment


    • #32

      Thanks for your suggestions, David and Stan.
      I am in the process of designing a new test bench with slider for greater focusing precision.
      I am using the aico 16mm F5.6 lens with BUC02. My lens holder is 22mm high from its base.
      What height are you using? This plays on the optical draw, I suppose.
      Do you finally use the capstan? and the thooted pulley wheel?
      I think I'll follow your setups which seem to be working fine right now (better than mine).
      I will also update the MSP to the latest firmware to be sure of the latest best version with the latest patches.
      Thanks again for your help.

      BTW :
      I think there is a way to integrate the little slider on Hawkeye by shifting it to the front
      but like you said david on an old post, maybe it's a lot of work for nothing.
      I will redo my tests on an external test bench.

      Comment


      • #33
        Hi Kamel. I used 20mm spacers with the BUC02 and the 16mm F5.6 Aico lens. At 20mm spacers used I see a bit of softness n the corners but that did not bother me too much because in the video and smaller screen that I normally use it is not noticeable. But as David suggested you should really run the bench test and see what spacing works out best for you. Sometimes it is better to go with the lower pixel resolution and better working distance for the lens. Once you blow that image up digitally and compare it with the image with higher pixel resolution it may actually look sharper in some cases.
        My personal opinion is that 20mm is pushing the limits of the 16mm lens and 22mm or even longer may be better.
        I managed to get the capstan to work. I already forgot all the details (its been a while) but the details are included in the manual. Let me know if you need any help with it. David also got it working. I believe he used the capstan without any sprockets.

        Comment


        • #34

          Thank you very much Stan,
          In fact I had put aside the project these last months and I am coming back to start again from the beginning.
          To save time, I preferred to ask you the questions directly than to search on the many pages of the Forum (which is a mine of information).
          I'm going to redesign a more precise lens mount and that's why I was wondering what height you are using today which gives the best result in optical printing. The information had already been given, it seems to me, but I have not found it.
          Attached Files

          Comment


          • #35
            Kamel, good luck finding focus.

            This is the lens
            https://aico-lens.com/product/16mm-f...ns-ach1622mac/

            I never paid any attention to these dimensions for Back Focus. It seems to be the spacing from the sensor based on sensor size. So all three sensors 2/3, 1/2, & 1/3 are all possible based on spacing. The lens part numbers are only different due to the aperture.
            Click image for larger version  Name:	aico.jpg Views:	0 Size:	127.9 KB ID:	32435 Click image for larger version  Name:	sensor_size_backfocus.jpg Views:	0 Size:	24.7 KB ID:	32436
            There must be some DOF for the rear element to cover different sensor distances.

            Any of these 3 projections are larger on our sensors. The BUC02 and the UX226 at 1/2.5 & 1/1.7.

            This is probably where my lens is now. This picture was from June 2020. I'm not going to open my Hawkeye to measure it.
            Click image for larger version  Name:	16mm_37ux.jpg Views:	0 Size:	228.7 KB ID:	32434
            I'm capturing and don't want any downtime. The V13 Hawkeye manual has dimensions for 12 & 16 for the BUC02 sensor.

            Comment


            • #36
              Thanks for the info David. Since my setup is open I decided to retest different spacing combinations with the BUC02.
              BUC02 with 16mm lens:
              20mm holder
              20mm spacers S8 rez- 2048x1526 overscan 1800x1300 cropped
              23mm spacers S8 rez - 1600x1200 overscan 1460x1100 cropped

              Here is the folder with test images:
              https://drive.google.com/drive/folde...pU?usp=sharing

              Looks like that the images with 20mm spacers are somewhat sharper indicating that the lens is still not reaching its limit as I earlier suspected. The improvement comes from higher pixel resolution. You can check the enclosed test images side by side.
              One downside is the scan speed which is 1 FPS max.

              BUC02 with 12mm lens:
              3/8" spacers:
              1280x1024 resolution
              Can run 2 FPS scan

              UX226 with 16mm lens:
              Standard holder + holder extension:
              5mm+8mm+1.5mm(adapter board) - 2500x1920 overscan - cropped a bit smaller

              Standard holder - no extension:
              8mm+15mm+1.5mm(adapter board) - 1600x1200 overscan 1400x1100 cropped




              Comment


              • #37

                Thank you very much Stan and David,

                David, glad you're in the capture phase.
                Your feedback is very interesting. Do you still have jitter?
                Do you finally use the capstan (with smooth or toothed pulley, at what position?
                What resolutions do your final images make?

                Actually, I have the BUC02 camera and aico 16mm F5.6 lens.
                I was not really satisfied with the results that's why I have to restart everything to start again on a good basis, and your advancements to all, will help me a lot.
                Thanks for all your testing and documentation.

                BTW :I think I found a way to integrate the little slider on Hawkey It will help a lot with the focusing.
                But Ihave to completely redesign the concept of the high front cover with an offset for the control switches.
                Still a little work, thanks again.
                (I will post my progress and of course the 3D files)
                Last edited by Kamel Ikhlef; April 19, 2021, 12:30 AM.

                Comment


                • #38
                  I find the focusing part to be pretty easy. With the 20mm spacers the 16mm lens sticks out enough that you can easily turn it in or out for the best focus. I adjust it until the image is sharp. Then back it off one way until it becomes a bit fuzzy and check how far I backed it off (can put a little mark on the lens). Then go the other way until fuzzy. Then turn it back again half way in between. I do this with the test pattern heavily zoomed in.

                  Comment


                  • #39
                    Kamel

                    Your feedback is very interesting. Do you still have jitter? No jitter. I have been running at the slowest speed, gate tension is adjusted, take up speeds are good. I have Hawkeye V12.

                    Do you finally use the capstan (with smooth or toothed pulley, at what position? Smooth with a rubber band in the second position.
                    Click image for larger version  Name:	editor-6.jpg Views:	0 Size:	97.1 KB ID:	32590 Click image for larger version  Name:	DSC06038.jpg Views:	0 Size:	91.3 KB ID:	32591

                    What resolutions do your final images make? Here are capture sizes below. In post I rotate the image, then I crop out the margins and perforations. Final render mp4 at 1440x1080.It looks good on 4k tv.

                    Click image for larger version  Name:	8mm.jpg Views:	0 Size:	117.6 KB ID:	32592 Click image for larger version  Name:	s8mm.jpg Views:	0 Size:	90.3 KB ID:	32593

                    The capture size could be larger, but I include both S8 & R8 in the field of view.

                    Updated
                    Just to be clear, I have been running Hawkeye at the fastest speed without jitter. Slow speed is for dark films (like the last one I ran), not all.
                    Last edited by David Brown; April 20, 2021, 03:40 PM.

                    Comment


                    • #40

                      Thanks very much David. You got a great workflow. I would like to arrive at your result, You and Stan (I am working on it at the moment).
                      The 2-position switch above the command switches intrigues me, what is this mystery?
                      For the capstan, very good idea of ​​the rubber around the pulley (
                      I don't know if this can solve the damaged perforation problems, but I will try).

                      what software do you use to resize your images? Me, I have davinci resolve but I do not know if we can make a precise cropping dimension type 1440x1080 for HD.

                      Stan, I tried your focusing technique but I'm not getting your results (I'm close! then
                      I continue my focus slider project)...

                      Thanks for taking the time to answer my many questions.
                      Last edited by Kamel Ikhlef; April 20, 2021, 04:37 PM.

                      Comment


                      • #41
                        Can you post the captured test image Kamel so that we can compare them. I crop the video directly in VirtualDub by using the crop filter. With the crop filter I make sure that the W and H are multiples of 4. Then run it through Video Fred. You can do additional cropping if necessary and resize.

                        Comment


                        • #42

                          Thanks Stan,
                          Unfortunately, I have my hawkeye unit completely disassembled because I am studying a solution to integrate a mini slider into it.
                          it will be easier for me (not sure that I can do it). I'm making good progress on my 3D design.
                          The last image I captured before disassembly is below.

                          https://drive.google.com/file/d/1K68...ew?usp=sharing

                          Comment


                          • #43
                            That is pretty close to my BUC02 test with the 16mm lens and 20mm spacers.
                            Click image for larger version

Name:	kamel-test.jpg
Views:	232
Size:	106.8 KB
ID:	32646

                            Comment


                            • #44
                              Looking at David's capstan design gave me an idea.


                              Use 2 of these to pinch the film. Believe some of the high end machines do it that way. One of them can be driven by a stepper. A flange can be added with the notches as a part of the motor encoder. Or the encoder can be another sprocket idler wheel. The idler wheel can be indented in the center so that it does not touch the image portion of the film. An MSP or Arduino can be used as a controller.
                              Pretty simple sw. Run the stepper until the signal is received from the encoder. Sent trigger to the cam. Pause and then run again and so on.
                              Bad perforations would not impact the operation. Too many perforations missing could cause issues with the encoder but making the encoder sprocket wheel large would mitigate that.


                              Comment


                              • #45

                                Yes, indeed close but less good. Your 63 lines are much better !

                                My images are generally less well defined than yours. I think that we can improve (the slide micrometer will be more precise but, I am perhaps not far from the limit of my material.

                                The last scan of the test chart that you posted a few days ago was not far from 80 lines (Very good),
                                I would like to at least achieve this result I think I'm on the right track.

                                For the pinch capstan, there is some elastic 3D filament which might be suitable.
                                Will it be necessary to drill the front trunk and provide a pressure spring?
                                In any case, a very good idea which appeals to me. Thinking...

                                Comment

                                Working...
                                X