USS Nautilus 3D
Collapse
X
-
Lots of good topside detail and smooth poly flow to minimize finishing work.
From the supplied finished print picture, the bow sonar chin mount fairing is inaccurate. (The leading edge should be vertical, rather than curved so much on the Y/ Z axis.) The NDD 3D Nautilus bow file shape is closer to reality.
For an RC application, this model's interior based on the parts of the exploded views, it would appear it needs extensive re-modeling to strengthen and align the hull sections instead of relying on small seating lips - especially with a horizontal hull cut bayonet-mount adaptation. A little work on the nose, and this has the potential to be a nice, large display model of the boat in its early period.
Thanks for posting!
CCLast edited by CC Clarke; 01-21-2022, 12:30 PM.Comment
-
This file is on sale (<$18!) for another day. I bit and bought it last night. There are a lot of files, so organizing them in separate folders is a big timesaver.
While there are a couple of detail flaws, (there were two major external revisions to this boat during its first couple of years of operation, and this model has them lumped together.) I've already begun repairing the most obvious flaws.
Detail-wise, it makes a good display model - the hull is very smooth, but has several million polys, so working with even one hull section within a 3D modeling program really bogs my computer down (and I have a decent system.) On the several parts I've remodeled, I was able to reduce the poly count considerably (99.9% on one!) with no loss of detail or faceting. I've taken some screen shots to show the difference for anyone wanting to try their hand at polishing this one up, and will post tomorrow.
Not recommended for filament printing; resin is the way to go with this one. I'll post back more detail as I dig into it.Comment
-
Das Boot: For resin printing, I have a Phrozen Sonic Mighty 4K. I'll be upgrading to an even larger 8k when the delivery times are reduced from 6 months. That will easily handle multiple (joined sections) simultaneously, eliminating seams and creating a solid part (though I still use 15mm frames when necessary and model others directly into the mesh. I'm moving away from FDM printing for all but simple things like supports and jigs for building. Even with really nice FDM printers, you're still going to have to deal with layer lines besides joint seams, which are inevitable. Resin makes the job so much easier, and for detail, it can't be beat unless you're machining.
There are a couple of geometry errors in the model and I haven't eye-balled the whole thing yet. The bow planes have a tear in the mesh that I repaired today. (Pics attached.) The top figure shows the elliptical hole, and the bottom figure is my finished repair - I added the cover bolts like the real deal (moored an hour away so I have plenty of late-model references.) I repaired the sonar fairing location on the bow for the Northern run. The model has it offset, but it needs to be on the centerline.
As I wrote in my last post, this is not a very good candidate for an FDM printer (like your Dremel.) I have access to professional-grade Raise3D printers, and I wouldn't waste the time or filament for a model with this much detail. Resin is a much better option for printing with < 50 micron resolution.
This is a display model - not designed for RC applications. A lot of modeling would need to be performed to get this where it needs to be.
The bow planes rigging uses a (vertical) linear slide. I animated it last night and it worked well, though it needs proper metal linkages and mounts modeled to align the parts.
CC
Last edited by CC Clarke; 01-26-2022, 07:32 PM.Comment
-
I've attached a few "Before and After" shots of some of the simpler Nautilus components that were quick to model. Note the polygon count in the Polygon Statistics panel. There is no loss of resolution with the re-modeled parts. The goal of 3D object modeling (as opposed to a straight CAD program which uses an algorithm to generate a mesh once the resolution (refinement in Fusion) is set is to work with all quads (four-point polys) and as few triangles (3-point polys). This makes it very easy to work with the mesh and change it if necessary without bogging down a computer, slicer, or printer.
A straight, cylindrical shaft doesn't need to be subdivided, and flat planes need very few polygons to define a shape. Ultimately, all polygons are turned into triangles for STL files.
Last edited by CC Clarke; 01-26-2022, 10:33 PM.Comment
-
The bow plane repair job didn't show up too well in the previous shots, so I zoomed in and tried to make it more visible.
One section had multiple points welded together into a single point, and the other section had missing polygons, showing the reverse side of the mesh. You can see through it because the polygons are single-sided, meaning you can look into the bow plane tear and see through the opposing side's polygons, but on the other side, they appear solid.
I'll show some more fixes, but don't want to get too bogged down with this model since I have others I need to work on. Oftentimes, it's easier (and a lot faster) to just re-build something from scratch rather than try to repair it.
Comment
-
While cleaning part of the sail up, I thought this tutorial might be of use to anyone contemplating cleaning up a mesh.
Polygon Reduction Technique.pdfComment
-
These are before-and-after sail renders after a modeling overhaul, which depicts the sail during the ships very early period. The polygon count was reduced by 25% from an initial million-poly start making it easier to work with. Changes include all masts and antennas, elimination and re-building of the port and starboard running lights, new railing, Deletion of the extruded hull number and awards, new front running light, and new exhaust diffuser.
This was a base mesh for two other versions, starting with the 1958-style, pre-polar run after MINSY re-worked the top and prior to the Essex taking a bite out of it. The top still needs to be rebuilt to match my drawings and reference materials.
Last edited by CC Clarke; 02-01-2022, 10:47 PM.Comment
Comment