Wolverine-Hawkeye Telecine

[David Brown]

Kamel,Stan,Bruce I just measured my motor shaft from the panel - 6.6mm or .26". I copied Bruce's picture to save time, it's exactly how I measured the length.

Kamel Don't redesign anything yet! I thought it would be better to move the new insert higher, equal to the original. There might be differences with the motor shaft lengths and the original insert might need to be lower. I don't know.

I think my problem could be prevented by using a vice or locking pliers. That would prevent the top and bottom surfaces of the support pulley from deforming when installing the insert.

Sounds like the jitter is coming from the takeup motor. If catching the film in a box solves the problem, what else would cause it?

I don't have any metered takeup on my V7 board and after I replaced the slip clutch in the take up, it never had enough force to cause jitter. Now I have replaced the original spring and find the takeup can pull the film thru the gate quickly when the claw is low. It's because those rotating pulleys move so easy! The original rubber coated posts stopped most of the force before it reached the gate. That and the gate pressure stopped the rest.

Is there anything else that could cause the jitter? Maybe just the gate pressure could make it more difficult for the claw (alone) to move the film.

Stan Is there any problem reversing the polarity on my motor? Would the encoder still function in reverse? I plan on leaving the capstan motor in the second position. It has solved my problems, still I will let it run a big reel and see if can sustain it's success.

[Bruce Davis]

Hi All, the motor shaft does have some end float, some is needed for expansion when motor warms but not as much. Checked the motor magnet and it has plenty of clearance to allow end float.

Removed the capstan motor to check the screws and found Stan had placed a 0.4mm thick washer so that screw heads did not make contact with the plate. Removed the washer, increased the counter sink so that screw heads did not protrude and re-assembled.

Hardly worth the effort all that was gained in shaft length 0.3mm (so from 6.5mm to 6.8mm). The measured shaft length with motor removed was 10.5mm, the printed mounting bracket is 1.75mm, the metal plate is 1.6mm, should be left with 7.15mm of shaft but in actual fact only have as mentioned above, all I have available now is 6.8mm.

Stan good to hear you are having success with v11.

[Kamel Ikhlef]

Thank you everyone for your tests and experiences.
Good News for V11, thank you Stan.
If you can improve the jitter on v11 this is very very good news for everyone,
because it is a recurring problem.

For the brass inserts holes, I lack space t o avoid the plastic deformations...
Maybe I can provide M2.5 or M3 plastic thread,
(without brass insert with longer screw for the upper part).
To be tested ... I continue...

[Stan Jelavic]

There is some jitter caused by the gate itself and/or the film but it is pretty small.
David, I believe the encoder should still be ok with the motor in reverse. To be tested.
Without a pulsed takeup jitter could be a problem. With the V11 I will activate the takeup while the claw is moving the film. So I do not see it pulling on the film while the claw is disengaged. I will test that today some time. It should make the takeup more smooth and possibly will not dislodge the film from the capstan. If it works then the first location should be ok.
I rechecked my shaft length calculations and looks like I forgot to subtract the backpulley to black panel clearance in which case it should be close to what David and Bruce observed. Will double check again.
Tightening the set screw does make the pulley crooked
https://photos.app.goo.gl/fwgkHqbKs6jPEF9M7
BTW - in the picture it looks like I am pressing on the pulley with the finger but I am not, just the angle of the microscope camera. I still believe one of the causes is that the shaft hole inside the pulley is too wide. Try inserting the motor shaft in without the set screw and wiggle the pulley. It is very loose. Can the part be done so that the hole is snug. Might help.

OK here are the latest measurements.

From Bruce:
The measured shaft length with motor removed was 10.5mm, the printed mounting bracket is 1.75mm, the metal plate is 1.6mm, should be left with 7.15mm of shaft but in actual fact only have as mentioned above, all I have available now is 6.8mm.

Shaft length from mounting bracket (Bruce) 10.5 - 1.75 = 8.75
Shaft length from mounting bracket (Stan's measurement) = 8.89

Shaft length from metal panel (Bruce) 8.75 - 1.6 = 7.15
Shaft length from metal panel (Stan) 8.89 - 1.6 = 7.29

Metal panel to set screw hole outer edge = 7.39 -- see picture below.

So, the shaft end should not be visible much when you look inside the set screw hole. And as they say the proof is in the pudding..
see image below.
The bottom line... if the shaft edge is not very close to the the hole end than something is misaligned on your setup.

[Kamel Ikhlef]

I am testing a new idea. M2.5 thread holes without inserts but M2.5x6 grub screws.
It works for me...
https://drive.google.com/file/d/1eQRAYBINB0QZjsDSykZWOKc0iDOEjNfj/view?usp=sharing

https://www.amazon.com/Yudesun-Hardw...8-1-fkmr0&th=1

BTW : I receive aluminum and copper filament this Thursday. I look forward to testing.

[David Brown]

That looks great Kamel! The location is lower and comes from both sides. Should fix both problems.

I'll just need to find (buy) some 2.5 screws.

The metal sounds interesting, I have printed wood. It actually smells like wood when printing! Good luck with that.

[Stan Jelavic]

Way to go Kamel. Looks like a pretty nice solution.
V11 udate:
Completed the tests today. All of them. Drilled a set screw hole on the opposite side of the existing one and added another set screw. It is possible to make the pulley straight by carefully tightening the screws but it is pretty tricky. But anyways with that the capstan works. The issues is that at this point it is required to keep the Wolverine rubber pulleys. Without them the pull of to takeup is way too strong and it causes jitter even when the takeup pulse is only 1 count long.
- three speeds 0.5,1 and 2 FPS
- rewind forward and reverse
- capstan monitor works in forward and reverse
- USB connection
- new frame sync
- capstan operation - the film does not slip off

Will post the videos tomorrow evening.

[Kamel Ikhlef]

Thanks everyone,
Stan, very good news for the V11 all tests...
I am very impatient for the schematic and MSP code update.

For rubber pulleys, we can print in rubber filament (TPU), keeping the curve to protect the films.
I will take the opportunity to modify the pulleys according to the dimensions reported by all.
(For a better alignment of the film). I will test ...

[Stan Jelavic]

The rubber pulleys should be fixed (not roller type). Bruce had some good results with the 2 set screw capstan. Not sure if he is using the new roller pulleys or not. There is another thing I observed with the takeup. The 12V drive is from the Wolverine power supply. Looks like they have a minimum voltage duration because no matter how short my pulse is (from the MSP) the takeup stays on for a while. Will hook up a scope to it to check. This could be redesigned to route the takeup control through the Hawkeye.
Here are the V11 design files:
https://drive.google.com/open?id=1-H...FT-2A_Y0Hm99ZA

[David Brown]

Stan Wonderful news on your V11 tests! I am anxious to get this going. Got an email from AICO, my lens will arrive Friday!

On top of all that it's March! I've been to Miami in August and March. A glorious time to be outdoors. Not in Utah, we're freezing, in Miami!

Has jitter returned with this newest version 11, or always been random? Does anyone have footage that will "jitter" on demand? I have not captured any "real" footage since I dismantled the Wolverine. I will be running a 400' reel soon just to test the capstan, though I will probably wait on V11. I replaced the original slip clutch spring into the take up. It is interesting that it is not as powerful as it was, but it's still too strong. The replacement was too weak to drive the capstan. The only adjustment is to shorten the spring so I'd rather do that with the final system. Looking forward to some jitter, the only time I had any turned out to be caused by the Super8 camera. edit: I should note that I usually restrained the take up with my hand having heard about jitter and the sound it made was disturbing!

[Stan Jelavic]

Hi David. Yes pretty nice weather in here this time of the year. But not in August unless you like plenty of heat and humidity.
As I suspected the takeup timing has a mind of its own. See the scope trace.

The blue trace is the control pulse from the MSP and the white trace is the 12V line going to the takeup motor from the Wolverine PS.
So, the takeup voltage has some minimum duration regardless of the control pulse duration.
Even with that the pulse width is much shorter than the claw cycle. It is around 200ms at 0.5 FPS. But looking at the takeup reel it looks much longer.
Could it be the motor inertia i.e. the motor keeps spinning for a while after the voltage is cut off? Maybe gear it down. Or maybe redesigned rubber pulleys as Kamel suggested.

[David Brown]

Stan it doesn't look like a stepper motor on the take up, and there could be a lot of torque needed inside the gear box.

So I could brainstorm here, but I'm sure you have already considered what ever I could suggest!

Knowing Kamel, he is thinking of rubber pulleys that rotate when friction is the key. Too difficult to add rubber strips that only touch the film edges?

Adding more friction to the film means the slip clutch will be slipping more.

Anyway, couldn't the voltage be dropped a little below 12v, or just skip a few pulses?

I can't even rotate the large exposed gear by hand. There must be a worm gear inside the box. Changing the gearing seems like a major project!

Originally, take up was belt driven. I can remember comments that the belt became greasy and started dissolving. That seems like a bad reason to add another motor. The driving gear is still visible on top of the claw's stepper motor. It would have been easier to sync speeds.

Just thinking out loud, returning to a belt drive wouldn't be easy!

Oh well, that's all I've got tonight.

[Stan Jelavic]

Good thoughts David. Maybe the easiest way would be to add a transistor to the board to drive the takeup. Since the MSP controls it anyways. I could just route the control pulse to the transistor. V12??? Will test it first.

[Brian Fretwell]

I'm just wondering when (like with steam locomotive restoration) someone will give up and build a better scanner by 3D printing camera circuit boards etc from scratch.

[David Brown]

Brian, I'm with you on that! Kamel is already printing all the parts including circuit board. He does it as an assembly model to check spacing. HEY, maybe it can work now with the new copper filament!!

Stan, not V12, all those finished boards will be sad. Do it as an in line module between the board and the motor?