Wolverine-Hawkeye Telecine

[Kamel Ikhlef]

No problem David. You will tell me later again. The idea of a manual frame-by-frame scanning is a very good idea, in case of blockage or too damaged film (yes, I have a lot og very damaged films, that's why the capstan can be very interesting) But I do not know if a "manual mode" is achievable (Stan can answer this question better).

[Stan Jelavic]

I am trying Kamel to meet different requirements but I am also constrained by the hardware capabilities of Hawkeye and limited space on the board. So far everything seems to work OK. Even the capstan. The only thing left is the USB connection and the power indicator.
The capstan can also work in 1 FPS mode. That can be achieved by setting the speed switch to slow and turning the turbo on.
David, the capstan motor is a small dc motor, not a stepper. It tugs onto the film and advances a very small amount until the damage is cleared:
https://photos.app.goo.gl/aFBfAx87o6nDbgp68
The stepper could be used but it would be some work to fit it in. Kamel's work also.
The 12V sonalert option is still available. There are two pads on the top of the board and it says SONALERT above them.
Regarding the two different circles David. You would have to produce images from both at the same distance by using a webcam. My guess is that the two images would be the same but the one from the 1/2.5" lens would be cropped. If so than it is just the mechanics of the lens which includes the housing that crops the picture and produces the smaller circle.

[David Brown]

Stan The capstan looks awesome! Glad to hear the alarm is still in place too.

To do single frame capture I am thinking of a secondary hand operated means to advance the film from an undamaged area. I don't see that capability incorporated into the Hawkeye hardware. Except for that interrupt switch. I suspected the capstan motor is not precise.

I will develop that idea. I see it as an external device. Based on my one time capturing without a triggered camera. I was able to stop the claw with a momentary switch, and release it after I saved the capture. That method should work with the the current system, we are saving stills based on claw position?

As I understand It, if the claw is stopped, the camera will wait until it resumes. All this would be to save a large length of damaged film. Not too common I would think.

Enough of that. The 2-16mm lens I have, seemed to be in different locations to get the same image. The 1/2.5" was too close to the gate.

I will set that up again and keep the pictures and positions next time.

Measuring can be tricky. I plan to measure from the surface of the webcam's circuit board to the end of the lens, and there to the surface of the gate.

Here is a shot that shows the lens length difference. (I just noticed that the threads are longer on the 1/2.5" the rear element is recessed !!)


Here is a view of the projected circles. I lined up the rear of both lens, not the front. (possibly a mistake now the difference is the threads)


Not so obvious here. The spots are focused on the paper. Left one is the 1/2".

[Bruce Davis]

Hi Kamel, Stan, David, not able to give much input on the technical side, just a comment on the capstan and damaged or broken sprocket holes. If the claw is stopped somewhere in its down position the film can easily be slid back and forth by hand (with gloves on) and each frame positioned reasonably accurately, all one needs is a method for the camera to be triggered....next frame etc until the sprocket problem is passed and then back to normal operation.
Thanks guys, regards - Bruce

[Stan Jelavic]

Thanks David,
Yes, the camera is triggered by the claw position. The MSP430 FW would have to be modified to provide a single frame. It should not be very hard to do. You would still have to advance the film manually (as Bruce suggested) because the perforations are damaged and the claw cannot advance the film. The film alignment will be an issue.
For the lens test setup the 2 lenses would have to be in focus i.e. you can have a bright object such as a back-light image and then compare the circles.

[Stan Jelavic]

Running through one of the reels last night at 1 FPS noticed the frames became smeared during dark scenes. Switched to 0.5 FPS and went OK. Light scenes again and 0.5 worked again. Still looking into it but got a hunch. The darker scenes need longer exposure and the claw starts pulling the film while the exposure is still active. And the root cause is POOR design. The claw pulls the film then goes down, back all the way to the left and then up... and this is the point when the trigger is fired. But then very short time after that the claw starts pulling the film and does not leave much time for camera exposure. 2FPS does not work at all, almost all frames are smeared.
But I think we have a workaround. We can sense the trigger. It is already used by the MSP for the takeup control. The MSP resets an internal counter every time the trigger is received. For 2FPS the counter max is 80 and for 1 FPS it is 160. A count can be checked, say if it is less than 10, the clock cycles can be stretched delaying the claw and allowing more time for exposure... That is in theory. Will test it and let you know.

[Bruce Davis]

Stan would it not make sense to have the trigger fire after the claw clears the sprocket hole and is in the down position on the right hand side before it moves to the left and up again. I gather by your comment you have no control over the trigger timing. Is there a sensor that can be relocated? Just some thoughts.
Regards - Bruce

[Stan Jelavic]

I had a similar idea Bruce. Unfortunately that would require a major rework. I currently do not have much control of the sync but there are alternatives. One possibility is to intercept the sync pulse in the MSP delay it and then route to the camera. There is one free output on the MSP (P2-7) and can be used for cam trigger. The delay can be large enough to put the sync at the point where the claw is down returning back.

[David Brown]

Stan, were you running a different reel than usual to find the darker exposures? The current lens I hope to buy is f5.6. That is an even longer exposure time for darker scenes.

The momentary switch I'm using on V7 is normally open. I have it plugged into J5/RUN which I thought to be the on off switch on the panel. It seems backwards, but I just put a clothes pin on button when I want it to stop the claw.

[Stan Jelavic]

I ran the second part of the reel with some dark interior scenes. Yes F5.6 will be even darker and will require longer exposure you are definitely right David.
Hopefully the fix that I am attempting will resolve the issue.
Mike never encountered the issue and he ran much more film than I did.
Grounded run switch is OFF position. Can change the logic. Not a big deal and can send you a new controller.

[David Brown]

You are too generous Stan! I was more interested to hear if that is normal or wacky. I should buy a proper switch, this one is off a Christmas toy's demo button. I'm waiting to see what you and Kamel come up with for the panel.

Scenes indoors are often too dark. Some scenes might still look bad if recovered. Changing the exposure likely uses the shutter or gain at some point. Increasing the brightness of the lamp seems like too much trouble.

If you can add in a timing delay, could that extend beyond the extra time in the claw cycle? Maybe for several seconds? Just a thought. I really don't want to introduce more work.

My dad bought an 8mm movie camera when I was a kid. In my entire life, I never saw him use any other camera. He even bought movie lights so all the stuff he shot indoors is well lit. However when my mother had the movie camera indoors, well we still haven't seen what was going on!

Good luck with this problem!

[David Brown]

Kamel

If you are looking for something to design and have free time, an M12 lens extension would help.

Someone on thingiyverse has tried but failed with the outside threads and the inside threads are very tight. The threads on the mounts you designed work perfectly with m12 lenses.

I can print it. Just repaired my 3d printer and it is working well.

https://www.thingiverse.com/thing:3817630 ( I printed this and the threads are not good)

[Stan Jelavic]

Still testing the frame sync and comparing with the V6 version board. The old circuit seems to be somewhat better based on a few tests. Will dig more into it this weekend.
Also noticed that Nick's Telecine does not work at 2 FPS whereas IC Capture works better at 2 FPS and the frames are ok most of the time except for dark scenes. I run in auto exposure mode but the gain is not auto and it is forced to the minimum. This ensures low grain but the exposure is generally longer.
For extended timing David I would have to redesign the hw. Not a big change but it would add lots of flexibility to the camera interface. Something to look into.

[Kamel Ikhlef]

No problem David, i will try several sizes. Tell me if you need a specific lengh size .
Anyway I'm stuck because I'm waiting for the advancement of Stan's tests to finalize the switch command front panel.
It is wiser to wait for the final v8 version with the improvements that Stan is making. Thanks to him.

[Kamel Ikhlef]

M12 Holder Lens Extension
Here the 3D Design (STL File) : https://drive.google.com/file/d/1w9B...ew?usp=sharing
I hope, it will works for you ! I think, the external steps are a little small . To test ...