<teepee>
ah, that thing. I did not know it can do fancy display too
<teepee>
it always wanted to print the models somewhere :)
<J2249>
easy to use for coloring and such - but with 60% chance the 3mf will have transformations applied - so when you print they are ok but your slicer will say something like "scaled 2150%"
<J2249>
but you get your thumbnail and license / author info ( but the 3mf is compressed in a way you can not change or add anything anymore)
<J2249>
(and it can not save an image of the scene .. so it is shift+win+s
<J2249>
)
<dTal>
J2249: oh my word, details pls
<dTal>
How did you design the blades shapes and how did you program them in openscad
<dTal>
and what's it for?
<J2249>
it is a profile lofted along .. it is for a extruder motion visualizer (prusa contest)
<J2249>
looks more usable .. are these waves not just for noise cancelling?
<dTal>
So they say - I think they're just there to look cool :)
<dTal>
I can actually see how it might reduce noise by offsetting the start of the pressure wave along the span, which means if you have straight stator blades the wave doesn't hit the stator all at once
<J2249>
the fillet on the hub is nice
<dTal>
thank you :D I'm very proud of it
<J2249>
is that a separate part where you moved a fillet along the profile and hub arc?
<J2249>
getting this with the sweep would be a lot of math
<J2249>
scopeuk just checked the most powerfull azimuth thruster have 22MW (4× at Queen Marry2) RR Mermaid
<J2249>
never thought a jet engine would have more power than the biggest ships
<dTal>
Ah no the fillet is done simply by swelling the profile during the extrude
<dTal>
still a lot of math, because obviously I don't have minkowski() - have to calculate the local derivate of the curve and offset it that way
<J2249>
but how is it curved like the hub. swelling the extrude would make it flat or?
<dTal>
the extrude is done before the conversion into polar coordinates
<dTal>
this is also why the tips are curved :)
<J2249>
ah that is interesting .. have never done that .. maybe i should write something to apply a polar distortion - sounds useful
<dTal>
it makes sense because the blade profiles are also calculated in terms of radial station
<J2249>
yes for rotating blades absolutely
<J2249>
cylindrical is probably enough .. can imagine where spherical would be useful except projections
<J2249>
and you need enough segments or need to calculate subdivision when bending a straight line
<dTal>
indeed
<dTal>
if you wanted to be extra fancy you could also consolidate segments which were curved and are now straight :)
<J2249>
i think CGAL is already doing that
nedko has quit [Ping timeout: 240 seconds]
<J2249>
at least you can not make a cube with more than 8 points (or square with more than 4)
<J2249>
i wonder if this wouldn't be a funny official SCAD module polar_extrude
nedko has joined #openscad
TheAssassin has quit [Remote host closed the connection]
<InPhase>
dTal: OpenSCAD can produce pretty much any shape computable, if you have the computer resources for it. polyhedron guarantees that, along with the general purpose features of the computing part of the language. It just takes some work to build up the right library pieces to make it easy.
<InPhase>
dTal: And in the vast majority of cases, the compute part is not the overhead, so it doesn't even matter that the interpreter part of the compute runs is below native performance.
<InPhase>
Generally computing with polyhedron is much faster than normal CGAL operations.
<InPhase>
A pretty decent fraction of my library work has been on just this. J2249 has taken up the same sort of philosophy, and has been more prolific than me in various options, made possible because J2249 saves time by not pressing the space bar. ;)
aiyion has quit [Remote host closed the connection]