Unfinished and untested 3d model of 120/220 developing tank

Hello, I decided to share my side project with the community.

The price of the print was estimated on 300 euro, so I gave up on it. According to the printers the screwing mechanism isnt optimised for 3d plastic printing and the whole thing is untested. Hence some tweaking is needed.

The spools should be able to take both 120 and 220 films. My idea was to load one film from the inside and one from the outside so I could have two 120 films on one spool.

The black part is a light trap based on a Penrose's unilluminable room. I have tested it with blender for light leaks and it should be light proof while allowing liquids to pass. Better print accuracy is needed because it is an optical device. There is some compensation in form of two such light traps merged together.


From materials you should pick something resistant to acidic and alkalic substances.

When exporting for 3d print do scale by factor of 1000, the exports in printers software behaves funky.

​

Download blender file here:

Cheers!

Took a quick look and I suspect that most of the design can be printed without too much trouble. The biggest challenge is going to be the actual developing spiral which will need to get split into two parts to print with FDM. Unfortunately I'm not skilled enough with blender to get it into a printable state. Right now none of the components are manifold and trying to get it exported as an STL results in my slicer splitting it into 88 parts.

thinkbrown said:

Took a quick look and I suspect that most of the design can be printed without too much trouble. The biggest challenge is going to be the actual developing spiral which will need to get split into two parts to print with FDM. Unfortunately I'm not skilled enough with blender to get it into a printable state. Right now none of the components are manifold and trying to get it exported as an STL results in my slicer splitting it into 88 parts.

Click to expand...

I can split it if you explain how I know nothing of 3d printing so we can merge our powers. Or I can upload the STL files

So essentially, 3d printing host software (commonly known as a slicer) takes a 3d mesh and turns it into a series of 2d toolpaths (slices). The software expects a printable object to be a single manifold mesh. I've included a couple examples of what happens currently when I export the blender objects as an STL. Essentially, to make the design printable, you'll need to join the various objects in blender into 4 final meshes that can then be printed.

I found this extension which is supposed to be able to help prepare models for printing: https://extensions.blender.org/add-ons/print3d-toolbox/. Unfortunately it seemed to remove a lot of the internal geometry of the light seal when I tried to use it, so I suspect I'm missing something.

thinkbrown said:

So essentially, 3d printing host software (commonly known as a slicer) takes a 3d mesh and turns it into a series of 2d toolpaths (slices). The software expects a printable object to be a single manifold mesh. I've included a couple examples of what happens currently when I export the blender objects as an STL. Essentially, to make the design printable, you'll need to join the various objects in blender into 4 final meshes that can then be printed.

I found this extension which is supposed to be able to help prepare models for printing: https://extensions.blender.org/add-ons/print3d-toolbox/. Unfortunately it seemed to remove a lot of the internal geometry of the light seal when I tried to use it, so I suspect I'm missing something.

View attachment 409595

View attachment 409593

Click to expand...

I have updated the website, there is now a new button where you can download the separate STL's for each object. Let me know if it works.

For the exports in blender you first apply modifiers on all the objects (its the wrench icon on the right panel), if there are some wierd objects like the text you need to go to object in upper left and convert to mesh. Then you trigger the x-ray view, select each group of objects which are supposed to be single stl, then go to export -> stl -> "selection only" and "scene unit" boxes checked, you might need to adjust the scale when exporting to other programs by some factor like m -> mm or m -> inch etc. The blender file is in meters.

Korbel said:

I have updated the website, there is now a new button where you can download the separate STL's for each object. Let me know if it works.

For the exports in blender you first apply modifiers on all the objects (its the wrench icon on the right panel), if there are some wierd objects like the text you need to go to object in upper left and convert to mesh. Then you trigger the x-ray view, select each group of objects which are supposed to be single stl, then go to export -> stl -> selection only and scene unit boxes checked, you might need to adjust the scale when exporting to other programs by some factor like mm -> m or inch -> cm etc.

Click to expand...

Getting closer. Currently each of the STLs appears to include the backdrop from the blender scene. When I split the file into objects in my slicer to remove the backdrop, all the letters were apparently not part of the same object. I imported it into blender, removed the backdrop, and this time the "make manifold" function gave me a single mesh that looks correct. I haven't tried with the other parts yet but this looks like we're making progress. I'm not sure if the scale is quite accurate.

thinkbrown said:

Getting closer. Currently each of the STLs appears to include the backdrop from the blender scene. When I split the file into objects in my slicer to remove the backdrop, all the letters were apparently not part of the same object. I imported it into blender, removed the backdrop, and this time the "make manifold" function gave me a single mesh that looks correct. I haven't tried with the other parts yet but this looks like we're making progress. I'm not sure if the scale is quite accurate.

View attachment 409615View attachment 409616

Click to expand...

Yeah I accidentally exported the backdrop I made for the render too, I will fix the uploads. The second Image looks accurate. The diameter of the whole thing is very close to 18cm. I might have miscalculated the lenght of the spiral maybe, I did it with some formula I found. Normal 220 wire is around 11cm diameter. Let me know if I should remove the letters completely.

Korbel said:

Yeah I accidentally exported the backdrop I made for the render too, I will fix the uploads. The second Image looks accurate. The diameter of the whole thing is very close to 18cm. I might have miscalculated the lenght of the spiral maybe, I did it with some formula I found. Normal 220 wire is around 11cm diameter. Let me know if I should remove the letters completely.

Click to expand...

Okay, if you're expecting it to be ~18cm that's fine. If you're going to keep the letters I'd recommend recessing them into the lid instead of extruding them. That way the lid can be printed on the top and won't need supports. Speaking of supports, what's the purpose of the little protrusion on the edge of the lid? That also gets in the way of printing the lid upside down

thinkbrown said:

Okay, if you're expecting it to be ~18cm that's fine. If you're going to keep the letters I'd recommend recessing them into the lid instead of extruding them. That way the lid can be printed on the top and won't need supports. Speaking of supports, what's the purpose of the little protrusion on the edge of the lid? That also gets in the way of printing the lid upside down

Click to expand...

I will remove the letters too. the protrusion on top was supposed to be a sort of handle to make opening and closing easier, I will remove it from the top too

Korbel said:

The black part is a light trap based on a Penrose's unilluminable room. I have tested it with blender for light leaks and it should be light proof while allowing liquids to pass.

Click to expand...

Theoretically, yes. In practice, this needs to be tested. I foresee two potential issues:
1: Diffuse reflections of the material may result in unwanted exposure inside the tank; the printed material is neither perfectly matt, nor a perfect mirror.
2: Fill and especially drain speeds may be an issue in practice.

koraks said:

Theoretically, yes. In practice, this needs to be tested. I foresee two potential issues:
1: Diffuse reflections of the material may result in unwanted exposure inside the tank; the printed material is neither perfectly matt, nor a perfect mirror.
2: Fill and especially drain speeds may be an issue in practice.

Click to expand...

I completely agree, I have tested it on 50% specular and 50% roughness material, but it is still a simulation.

The holes have a total area of 11cm^2 and accoring to AI it should do around 0.8 liters / s

the red bowl is around 3 liters volume so a rough estimate would be 3-4 seconds speeds with the spools, I dont think paterson tank of such volume is faster, but I dont have it, just seen some videos.

Korbel said:

The holes have a total area of 11cm^2 and accoring to AI it should do around 0.8 liters / s

Click to expand...

I don't doubt the holes are big enough in principle. But I doubt AI takes into account real-world dynamics of a fluid. If you flip a bottle of water upside down, what happens? I think you'll see similar effects in this setup. The 'trick' of the Paterson, Jobo etc. tanks is that the light trap is circular and goes all around the lid, basically. This means that whichever way you tilt/flip the tank, it'll always drain pretty fast since there's always a way the air will be able to get in. In your design, orientation will affect drain rates more strongly. Same for fill rates; your design is liable to 'locking up' if you try to fill it too fast.

Mind you, it may still work OK, perhaps with same care/taking into considerations some best-use practices.