Here is a Factory File containing this part:
What I am doing now is trying out different settings for extrusion width, Solid Diaphragm every XXX layers, etc.
Top layer strangeness
I am experimenting with a simple cup shape to determine how to make objects that are watertight - meaning they will hold water and not leak. Here is an image of my test part:
The part is a fairly simply cup shape with a flat bottom that has an upper (top) and lower (bottom) surface. When I look at the individual layers forming the bottom surfaces they all look good, except for the top 2. Here is a pic of the bottom's top surface layer:
I don't understand why the squarish shapes are there and I think they will allow water to seep through in between the individual filament traces. Up to now all the parts like this I have made have leaked water and my sense is it is because the adjacent traces of plastic are not bonding completely to each other.
Here is a Factory File containing this part:
What I am doing now is trying out different settings for extrusion width, Solid Diaphragm every XXX layers, etc.
The part is a fairly simply cup shape with a flat bottom that has an upper (top) and lower (bottom) surface. When I look at the individual layers forming the bottom surfaces they all look good, except for the top 2. Here is a pic of the bottom's top surface layer:
I don't understand why the squarish shapes are there and I think they will allow water to seep through in between the individual filament traces. Up to now all the parts like this I have made have leaked water and my sense is it is because the adjacent traces of plastic are not bonding completely to each other.
Here is a Factory File containing this part:
What I am doing now is trying out different settings for extrusion width, Solid Diaphragm every XXX layers, etc.
Re: Top layer strangeness
There are a few inaccuracies in the surface of the inside of the bowl that are being picked up by the slicer as miniature indentations...which can be seen by the two sets of weird rings in the top layer.
I've fixed the bowl shape by aligning all the nodes on the inside surface and after reloading in S3D, the slicing is perfect!
Revised shape file attached. I've zipped it because for some strange reason, the .stl file is larger than the factory file I downloaded!
Sceptics would instantly jump at the slicer software being at fault.......but this is a prime example that the slicer is trying to do its best by slicing what it finds!!!
I've fixed the bowl shape by aligning all the nodes on the inside surface and after reloading in S3D, the slicing is perfect!
Revised shape file attached. I've zipped it because for some strange reason, the .stl file is larger than the factory file I downloaded!
Sceptics would instantly jump at the slicer software being at fault.......but this is a prime example that the slicer is trying to do its best by slicing what it finds!!!
- Attachments
-
- watertest_fixed1.zip
- (2.77 MiB) Downloaded 217 times
Doug Kightley
Volunteer at the National Tramway Museum http://www.tramway.co.uk
Railway modeller and webmaster at http://www.talkingtgauge.net
Volunteer at the National Tramway Museum http://www.tramway.co.uk
Railway modeller and webmaster at http://www.talkingtgauge.net
Re: Top layer strangeness
You are correct - the revised STL slices perfectly. Thanks.
But what is it you actually did to fix the file? What does "align nodes" mean? ANd how did you identify what the problem was?
I generated the original STL straight out of Rhino and I probbaly ran it through the 3D Builder repair process. Everything looked good to me until I did the slicing. If there is a better way to identify subtle problems with parts like this I'd sure like to know what it is.
But what is it you actually did to fix the file? What does "align nodes" mean? ANd how did you identify what the problem was?
I generated the original STL straight out of Rhino and I probbaly ran it through the 3D Builder repair process. Everything looked good to me until I did the slicing. If there is a better way to identify subtle problems with parts like this I'd sure like to know what it is.
Re: Top layer strangeness
My investigation went as follows:
Firstly, I downloaded the factory file and looked at the preview to see exactly what you were talking about. To exclude slicer issue, I rotated the bowl and moved it around and looked at the sliced product. The errors moved around with the bowl.....so there had to be something wrong with the shape!
I exported the shape to an .stl , and imported into my CAD package to view the mesh:
I highlighted the nodes in the mesh in the area of the surface of the bottom of the bowl. You can see some highlighted in the image above.... a ring around the outside surface and all of the nodes in the bottom surface of the inside of the bowl.
Taking a closer look:
Notice there's a few nodes that are slightly lower than the rest...it's these that were causing the micro-hole in the slicing. All I did was to simply select the nodes right across the whole shape and align them vertically. Sort of ironed out the irregularities! 
The only complication was that when I exported the mesh back out as a .stl file and loaded into S3D, it was now 1/10th of its original size!! So I had to scale it by a factor of 10 and exporting back out as a .stl file for uploading.
Firstly, I downloaded the factory file and looked at the preview to see exactly what you were talking about. To exclude slicer issue, I rotated the bowl and moved it around and looked at the sliced product. The errors moved around with the bowl.....so there had to be something wrong with the shape!
I exported the shape to an .stl , and imported into my CAD package to view the mesh:
Taking a closer look:
The only complication was that when I exported the mesh back out as a .stl file and loaded into S3D, it was now 1/10th of its original size!! So I had to scale it by a factor of 10 and exporting back out as a .stl file for uploading.
Doug Kightley
Volunteer at the National Tramway Museum http://www.tramway.co.uk
Railway modeller and webmaster at http://www.talkingtgauge.net
Volunteer at the National Tramway Museum http://www.tramway.co.uk
Railway modeller and webmaster at http://www.talkingtgauge.net
Re: Top layer strangeness
Fascinating - many thanks. It never occurred to me to look at the actual mesh resulting from the Rhino Export function. Of course doing that makes perfect sense to me now. So I will go back to my original file and see if I can reproduce what you did.
Re: Top layer strangeness
Well, a little investigation yielded conclusive results and what seems to be a workable solution.
I imported my original STL file into Rhino and zoomed in to the area between the bottom and side surfaces. As you noted there were some weirdly placed triangles in several places. These were clearly the same anomalies you noted and fixed.
As you may know, Rhino uses NURBS surfaces for it's models, and these are useless for printing. So in order to produce a valid STL file Rhino has to first create an internal mesh. Because most of the parts I design are large I thought I could minimize exported file size and print time by specifying a meshing tolerance of 0.01 mm.
Apparently doing this was a mistake, because when I changed the tolerance to 0.001 mm the anomalies previously generated no longer occurred. So I'll stick with that from now on, regardless of the size of my part.
Thanks again for your help with this.
I imported my original STL file into Rhino and zoomed in to the area between the bottom and side surfaces. As you noted there were some weirdly placed triangles in several places. These were clearly the same anomalies you noted and fixed.
As you may know, Rhino uses NURBS surfaces for it's models, and these are useless for printing. So in order to produce a valid STL file Rhino has to first create an internal mesh. Because most of the parts I design are large I thought I could minimize exported file size and print time by specifying a meshing tolerance of 0.01 mm.
Apparently doing this was a mistake, because when I changed the tolerance to 0.001 mm the anomalies previously generated no longer occurred. So I'll stick with that from now on, regardless of the size of my part.
Thanks again for your help with this.
