Defeaturing of slightly curved geometry in slicer to enforce planar top layer  

RE: Defeaturing of slightly curved geometry in slicer to enforce planar top layer, Example Picture

Here I add one example picture to make more clear what I mean:

RE: Defeaturing of slightly curved geometry in slicer to enforce planar top layer

Hi Nils, 
it would have been better to include the project file, including the STL's (suitably Zip Compressed) for us to play with

could you use the 'Cut tool'  to cut the model a tiny bit below full height, to force a smooth planar surface?

regards Joan

RE: Defeaturing of slightly curved geometry in slicer to enforce planar top layer

Hi Joan,

thanks you for response!  Below you can find the files.

Cheers, Nils

  blockC

RE: Defeaturing of slightly curved geometry in slicer to enforce planar top layer

Block_C JT
Is this OK?

Joan

RE: Defeaturing of slightly curved geometry in slicer to enforce planar top layer

Hi Joan,

this looks great, Thank you very much!  

But, how did you do this? This looks like cleaning up geometry in another software and not like only changing some slicer settings, right?

Or did you use prusaslicer do get it like this?

Thanks again!

Cheers,

Nils.

Nils

RE:

Hi Nils, the initial picture did not show all of the model geometry. 
If the model had simply been a round cornered tapered sided pad, with a slightly lumpy top, slicer could be used to cut the top 0.75 of a millimeter off, and then the part could have been re sized in just the Z axis, to restore height however the actual model had a second mound towards one end of the top surface. which made life difficult.

looking at the STL, it was supplied, tilted in space (In two directions) and the bottom, middle and top surfaces were not parallel, 
The small  amount of apparent roughness on the bottom layer seemed to sort it's self out, when placed on the slicer build plate, leaving the mid layer, largely parallel, apart from surface roughness, but the top layer seemed to be tilted as a couple of degrees angle to the main body.

I found it easier to work with the version of the STL, that was inside the .3mf file, than the original STL, simply, because it hac been lowered to the build plat, and that had removed the major tilt along the X axis and the minor tilt along the Y axis...   (This is difficult to do in My 3D design package)
So I Exported the STL from Prusa slicer and imported the same STL into my 3D Design package.

once the model was inside my 3d package, I created a thin plane sheet, 300mm x 300mm x 1mm (Lazy measurements to ensure the sheet covered the model)
I placed this sheet on the lower surface of my model build area and then raised it slowly up the model, watching the plate, obscure the model as it raised, I noted that the Mid level  was largely parallel to the model underside but there were small un even areas on the top face of the middle level.    There was one, small artefact on this face, which looked like  it may have been intended to be a triangular pointer . 
Not knowing whether it was supposed to be there or not, I Noted it's presence, and continued... (If it is important, we can re create is again, later)
I continued to raise the flat sheet up to the top of the model and noted that the top of the model was tilted a couple of degrees, relative to the bottom and middle 'mean' layers (I use 'Mean' to describe the surface which would remain, if I removed the surface artefacts. )
I didn't know if the top surface was supposed to be tilted, or parallel to the base, however you had stated the desire for the top surface to be laminar, and a slope, cannot be laminar on a traditional 3D printer. 

this effectively mapped out the work for me to do...   
In hind sight, I would have done the task differently. however. 
What I Did do, was to lower the 1mm flat plate down the model until the underside obscured the centre parts of the Mid level surface, but retaining most of the Rounded edges, as if by Magic, the top surface of the 1mm flat plate, covered pretty much all of the surface roughness on the top of the top surface, leaving the dome, to stick out over the top. 

I then executed a CUT Command to cut a 1mm slice, out of the model, which removed the mid surface roughness and a tiny bit of the bottom of the mound...  

At this stage, the model was in two parts, held in space, relative to each other... BUT The 3D package considered them to be one element...
(this was not ideal, but it was where I Was... ) (I could have split the components but in this case I simply exported them to an STL file called Block_C split

I imported the STL, into Prusa Slicer, and then used the 'Split to Objects' tool to split the parts...  
AND...   the top dome, fell through the main model, to end up sitting on the floor...    which was not ideal. 

Prusa slicer does not like you trying to 'balance' items in 'mid air' so raising the dome was a task for later. 
whilst the dome was on the floor, I Could clearly see the tilt in the dome top surface. Using the rotation tool, I changed the angle of the model, relative to the build plate, until it looked level. 

then I selected the dome part, right clicked and used the 'ADD PART' menu option to add a small cylinder, to the dome. 

I placed this cylinder out of sight inside the main part of the model, resting on the base of the build plate, this gave me the option to move the dome relative to the cylinder (the cylinder was effectively sacrificial) 

NOW I was ready for the finishing bits
I Knew that I had removed approximately 0.75mm from the main body height, so I Unlocked the scaling tool, in the right hand window in Prusa Slicer, and I Scaled the main body 0.75mm taller ('Z') leaving the 'X' and 'Y' dimensions unchanged.

now for the hard part, I Scaled the dome in the 'Z' direction a little, and adjusted it's position  in the Z direction. until the top surface was about the right height for the final model, and the 'Skirts' of the lower flared side were below the surface of the main part of the body... 
I wanted the minimum change in the vertical height of this dome, that would allow me to hide the skirts and place the top, in the best position...   (Changing the height of the taper sided section of the model, changes the resultant angle of the slope, and I wanted to minimise the changes)

Once everything was in position, I Selected all parts, within Prusa Slicer, and used the merge tool, to make them one piece, which then allowed me to slice the model, successfully

regards Joan

Model project file, in Zip File below

RE: Defeaturing of slightly curved geometry in slicer to enforce planar top layer

Hi Joan,

thank you very much for your detailed answer. I now understand the approach.

There are really some expert tricks  you applied here. 

Best regards, 

Nils.

RE: Defeaturing of slightly curved geometry in slicer to enforce planar top layer

Glad I could help. 

regards Joan