Move outer cylinder face on an object

[peter_spears]

I'm stuck on a workflow, and wonder if someone out there can help me. The test print I'm working with is quite simple: A cylinder with 60mm ID and 2mm walls, 20mm high. I have some text on a sketch, which I have used to deboss the outer surface of the cylinder by 1.5mm.

This works nicely. It leaves me with the text debossed with a thin wall of 0.5mm, and this creates a nice lighting effect when printed with white PLA. The problem is that it does not print the 100% overhangs cleanly. What I would like to do is move the outer face of the cylinder up by 0.5mm, to create a printable overhang.

I have tried using the Move tool, selecting Faces and then selecting the outer cylinder face, plus the holes in the letters, and moving them. It seems to think that it moves them, but doesn't do anything. There is no error, either - just does nothing. I would have expected it to move the selected faces, and make what length and angle adjustments it needs to in the adjoining edges to achieve the move.

Any thoughts on how to move past this?

[g-andresen]

Hi,

Either add supports (slicer) or ramps (fusion > chamfer)  for the overhangs.

Please share the f3d file with solid body for reply

File > export > save as f3d on local drive > attach to post

günther

[peter_spears]

Thank you

Supports are a possibility and I've got some good results with them, but I'd prefer to avoid them if I can. This is a proof-of-concept for a design I'm working on that will be considerably bigger and more intricate, so there will be a lot of scaffolding to print.

I have tried chamfers. They work in most cases, and give me the option of only chamfering the faces that are unprintable. The problem is in the cases where I can't. In the attached model, the first curve on the top of the 'm' won't chamfer no matter what I do, and it won't print.

Ideally, I'd like to grab the face on the outside of the cylinder, along with the islands inside the letters, and move them up by 0.5mm. Fusion will let me select them, but the move does nothing.

I'm looking into non-planar printing, too, which would also resolve this. It's a steep learning-curve, though.

Peter

[etfrench]

Have you tried printing supports with a different filament?  Petg and pla work well for this.  If you print in ABS, you can use HIPS which dissolves in limolean.

[peter_spears]

Thank you for your suggestion. This would be possible solution, in that the PoC I did previously used supports, and left me a lot of work to clean up afterwards (even with organic supports). However, there are two problems with this: Firstly, my printer only has a single extruder, and I don't have the bandwidth at present to incorporate another one in my current design, and secondly, printing supports takes a significant amount of time and material, and I am looking for ways to print this model without supports.

I should add here that the printer is not stock anything. I designed and built it myself, based on the MPCNC project. The closest I've found on the market is the FL-SUN V400. I'll attach a picture of the printer, and of the test print that shows the issue I am trying to address.

As I mentioned yesterday, I'm looking into conical slicing, which I'm hoping will address this issue. Given the design, I think this approach will work, but it is not built into any slicer yet so the workflow is considerably more involved. I still think having the ability to move the outer face of the cylinder is my best option.

[etfrench]

Loft may work. Try embossing larger letters, then surface workspace lofting between the bottom edges of the embossed and debossed letters.  Embossed letters can be removed with a split body command.

[g-andresen]

Hi,

I created a conical (-9 deg) text on a sheet metal body and embossed it onto the ring via the refolded body.

günther

[peter_spears]

Ah, thank you, Günther. I didn't think of using sheet metal. This opens up all of the tools that work well on a plane. The difficulty with this is closing the seam afterwards, so it it not visible in the print.

I have managed to get the desired effect by exporting the object as an STL, importing it into Blender and then doing the cleanup and transformation in there. This at least achieves what I was trying to do, although it would be nice to be able to do this in Fusion.

I have also tried, and will continue to work on, non-planar conical slicing. The back-transformed model looked fine in the gcode viewer, but the printer didn't like it and on inspection, it seems to have done something weird with the extrusion rate. If I can get this working it would be ideal: I'll be able to print the horizontal overhangs without support. I'm following the guide offered by CNC Kitchen.

[g-andresen]

Hi,

---

@peter_spears  schrieb:

The difficulty with this is closing the seam afterwards, so it it not visible in the print.

---

1. This was caused by an incorrect positioning of the sheet metal body. Here I have moved the sketch to a different level.

2. It is also important to use the smallest possible nozzle size and the smallest possible layer height.

günther

[etfrench]

I printed this on a Bambu X1C with 0.12mm layer height, no supports, and a 0.4mm nozzle.

The black spot in the 'a' is a bit of black filament from the previous print.

p.s. A print on my delta without supports would look more like your print.

[peter_spears]

This is a clever and simple solution to a complicated problem. Thank you. So just subtract the refolded cylinder from the original model.

The only issue with this is that I need the angle to be consistent. Extrude allows you to set a taper angle, such that the extrusion tapers inwards by 9d - the top tapers down, and the bottom tapers up. I need the angle to be a consistent 9d. I could probably do this by copying the sketch and using Loft, but I would need to loft every letter and island separately, because the Loft tool only allows you to loft one profile at a time.

What I really need as an Extrude tool that takes an extrusion angle, rather than a taper angle. Is it possible to get Fusion to do this? 

[peter_spears]

Wow! That's an amazing result. And you did this without modifying the model?

What is the difference between the Bambu and the Delta, then, that yields such vastly different results, and is it possible to get my Delta to behave the same?

I haven't tried reducing the layer height that much. I'll give it a try and see how it goes.

[etfrench]

It was printed using the default settings for PLA.  The Bambu is a corexy printer and runs at much higher speeds than a delta or a bed slinger. Infill is 180 mm/sec, inner walls are 150 mm/sec.  Outer walls run slower for smoother perimeters at 60 mm/sec.  I ran my delta at 100 mm/sec which was significantly faster than bed slingers at 60 mm/sec. 

I think the design of  your delta is quite good for an inexpensive 3d printer.  3d printed carriages running on non-precision bars won't be as smooth as high quality linear rails.  That is not to say your printer isn't capable of making good prints.  The quality of the X1Cs prints are easily ten times better than my delta prints.  I'm not sure there is an easy  way to add more rigidity to  your printer.  Side panels on two sides would help.  PTFE tape in the arm sockets may reduce some friction. 

Take a look at the Voron corexy DIY printers.  They should be capable of matching the X1Cs quality.

[g-andresen]

Hi,

The pictures in your post 5 show that the quality (visible layers) is "not very good" even in unproblematic areas and at the transition from the lower edge.

günther

[peter_spears]

Thank you - that's what I was aiming for, an inexpensive build with okay quality. I've done two prints of the final design, one at 150mm/s and one at 75. The slower one is MUCH better, so you are absolutely right about problems with rigidity.

The original plan was to fully enclose the cavity with plywood, with the intention of being able to heat cavity and be able to print ABS or nylon. However, plywood was a terrible choice for this because it warped terribly, so I removed it. I'll still enclose it eventually, but in the meantime, adding solid diagonal braces to two sides will at least help reduce the wobble. The conduit is also considerably more flexible than an extruded linear rail, but I think that's something that I'll need to live with, at least for this design.

Thank you for the pointer to the Voron CoreXY DIY printers, too - I never realised that the Voron designs were open source. I will explore these in more detail. PTFE tape is also a good approach to reducing friction in the sockets. I've also considered printing the sockets with carbon PLA, which has lower friction that standard PLA.

[peter_spears]

Yes - thank you for pointing this out. As etfrench pointed out above, there are some small changes that can make some big improvements with rigidity, so I'll address those first.