Large reels?

[robustini]

Thank you for the great project!
I didn't quite understand from the documentation how it handles the two NEMA17s, i.e., how it makes the film always stay tightly wound on the output reel.
I understand that every single frame is detected by an algorithm that intercepts the drag hole, right?
Other than that, I have several 20cm reels, wasn't a larger frame designed?
Regards,
Marco

[stuartpittaway]

The code keeps track of how much film has been used and tightens up the slack in the reels.

You could fit larger reels, would just need the arms lengthening

[robustini]

Thank you for your reply!
I have pretty much everything at home to try your project because I think it is very interesting, I have RPI 5, RPI HQ Camera, the 100x lens, Nema 17 steppers, some compatible Marlin boards with TMC2208 drivers, and 6000K 6W lamps.
But modifying the frame for larger reels is not trivial since the one wrapping in the project is vertical and not diagonal like the one on the left.
Surely you are better at this than I am since I only use Fusion 360, could you make this modification?
Thanks!

[robustini]

Also I see from your movie that there is a particular molded stand to attach the 100x microscope lens but I don't see this printable support in your project.
But is this repo updated with everything needed as printable parts?

[stuartpittaway]

Yes the repo is correct. I believe some of the parts are toggled off/not visible in freecad - just need the visibility toggling back on.

I've not touched this project for a few years now, and don't have the time to amend the machine for a larger size.

You may be able to take the 3d parts into fusion 360 and stretch them?

[robustini]

Work in progress.
I just have to divide it because in one piece it is too big.

[robustini]

@stuartpittaway please check because the support for the 100x microscope that you see in the video does not exist in the project you posted, or at least I don't see it in FreeCAD.

[robustini]

The code keeps track of how much film has been used and tightens up the slack in the reels.

If you use the molded reel from the project maybe yes, but with another one that maybe has a different internal diameter I don't think it would work well, right?

[burgsth]

Could you please share your 20cm reel version? Maybe there is also an additional deflection needed ob the input side caused by the significant different angle if the spool is getting to its end.

[robustini]

Could you please share your 20cm reel version? Maybe there is also an additional deflection needed ob the input side caused by the significant different angle if the spool is getting to its end.

I had already corrected it but thanks for bringing it to my attention.
I don't know if I will continue with this project, I don't really like that the reel that retrieves the film doesn't have a control of when to activate and stop, such as with T-Scann 8 is expected.
Also, the project on GitHub does not have some parts I need.
I immediately liked the frame stop system based on the drag holes, but there are other things that puzzled me unfortunately.

[stuartpittaway]

@stuartpittaway please check because the support for the 100x microscope that you see in the video does not exist in the project you posted, or at least I don't see it in FreeCAD.

Look in the file CameraMount-RaspPiAdvanced.FCStd for the Rasp PI advanced camera parts.

[stuartpittaway]

I don't know if I will continue with this project, I don't really like that the reel that retrieves the film doesn't have a control of when to activate and stop, such as with T-Scann 8 is expected.

I didn't find it a problem on the dozen or so reels I scanned - although they were all the smaller reel sizes.

The lower motor pulls the film through, the source film is simply free spinning.

Once the code has determined there is some slack in the film, the take up reel is rotated to pick up the slack - not perfect, but worked okay for what I wanted to do.

You can obviously sit and watch the whole process and manually turn the pick up reel if you wish instead.

[robustini]

Once the code has determined there is some slack in the film, the take up reel is rotated to pick up the slack - not perfect, but worked okay for what I wanted to do.

You can obviously sit and watch the whole process and manually turn the pick up reel if you wish instead.

Certainly with small reels it can work, but with large reels in my opinion a huge error will be generated.
It is unthinkable to stand in front of the scanner for many hours to retrieve the film manually.
The 3d printer board has many connectors for mechanical or optical endstops, it would be useful to implement something that activates the stepper of that reel only when really needed.

[stuartpittaway]

Agreed, plenty of inputs on the 3d board for some sort of sensor.