ghane changed the topic of #solvespace to: SolveSpace--parametric 2d/3d CAD - latest version 3.1 - http://solvespace.com - https://github.com/solvespace/solvespace - This channel is being publicly logged on https://libera.irclog.whitequark.org/solvespace
<ghane> ddevault: when creating a new workplane you can either use a point, in which case the cooridnate axes are used for the plane or select a point and two lines, if you add more construction lines to your original sketch I think you could do that
<ddevault> I did add a few more construction lines but I can't seem to get a plane which is tangent to the cylinder
<ghane> otherwise you could just create one spoke that aligns with a plane (and is tangent to the cylinder) and revolve it
<ddevault> for example, if you look at this model
<ghane> sorry not revolve "new group step and repeat rotating"
<ddevault> if I select the nearest point, and the two lines across the cylinder at either end of the line through the middle, I would assume I get a workplane tangent to the cylinder
<ddevault> but nope
<ddevault> ah actually that line is not tangent
<ddevault> but I did create more construction lines which were tangent to no avail
<ddevault> lemme play with it some more
<ghane> :)
<ddevault> another solution that occured to me this morning is to just add spokes which run the full length of the cylinder and then carve out pegs with a difference drawn over top
<ddevault> so here's another example: https://l.sr.ht/JedG.png
<ddevault> if I choose the point on the cylinder nearest the viewer, and the two lines of the reference cuboid that form a tangent plane, I don't get a tangent workspace
<ghane> that sounds strange, but if you select the point nearest to the viewer in that case you should be able to get a new workplane aligned to the coordinate axis that should be tangent?
<ddevault> I was able to do this by creating a workplane from the point and two perpendicular lines on the tangent face of the reference cuboid
<ghane> ah, you selected parallell lines?
<ddevault> yeah
<ddevault> what was that you were saying about step repeating? can I repeat this cylinder on 6 sides?
<ddevault> messed with it but not sure what things to select to get the appropriate configuration
ghane has joined #solvespace
ghane has quit [Changing host]
<ddevault> got it https://l.sr.ht/1119.png
<ddevault> thanks ghane!
<ghane> ddevault: ah you were faster than I was :)
<ghane> nice
<ddevault> solvespace is a wonderful tool, really appreciate the work y'all do here
castaway has joined #solvespace
<ddevault> uh
<ddevault> the outer face of my core cylinder appears to have disappeared
<ddevault> and an inner face of this spoke attachment is also missing
<ddevault> the fuck
<ddevault> simplified it, got here: https://l.sr.ht/jpne.png
<ddevault> I think what I needed is assembly, not union
<ddevault> not sure why though
<ddevault> this whole part is eventually going to be diff'd out of a larger cylinder
<ddevault> hope that works
<ghane> Oh, yes that happens sometimes... the solver calculates something incorrect, think it usually happens when curved surfaces are intersecting in some odd way
<ghane> a workaround (which might hurt the resolution) is if you select the mesh in the property browser and check "force NURBS surfaces to triangle mesh"
<ddevault> I got my model working now without a workaround, by using an assembly extrusion rather than a union
<ddevault> no complaints from solvespace https://l.sr.ht/O8nE.png
<ghane> ah yes that works too
<ddevault> one weird degree of freedom in the rotate group, not really sure what that's about
<ddevault> okay, so I can't really figure this out
<ddevault> given this part: https://l.sr.ht/PYzo.png
<ddevault> how do I subtract it from a larger cylinder
<ghane> I think the easiest way is to create a new file with the larger cylinder
<ddevault> then?
<ghane> Then link it using New group linking / assemling file (Shift-I)
<ghane> and use boolean to subtract (if I am not mistaken, I only have Solvespace 2.3 on this laptop)
<ddevault> that almost works, but I get some weirdness
<ddevault> I get some missing faces and the cap of the cylinder does not subtract from the other cylinder
<ddevault> I would also like to get rid of the degree of freedom in my rotate group but no clue what causes it
<ghane> yes, I think that missing faces is the same (rounding?) error... try force triangle mesh for it
<ghane> not sure what that degree of freedom could be
<ddevault> does not really fix it
<ddevault> just makes it differently broken
<ddevault> w/o triangle mesh: https://l.sr.ht/XL3k.png with: https://l.sr.ht/acd0.png
<ddevault> I also noticed that exporting the wheel model as an STL brings up errors about edges
<ddevault> so the problem may be there first and foremost
<ddevault> these are the "problematic edges" https://l.sr.ht/EmSU.png
<ddevault> I think I'm running into these problems https://solvespace.readthedocs.io/en/latest/issues.html
<ddevault> but the suggested workarounds don't seem to do much for me
<ddevault> figured out how to get rid of that dof btw
<ddevault> needed to constrain an angle of one of the copies
<ghane> It seems you have redundant constraints on that one?
<ddevault> not according to my solvespace
<ddevault> version 3.1
<ghane> Yes, I should check later when using correct version :)
<ddevault> well
<ddevault> remade the model from scratch and got it to work
<ddevault> ¯\_(ツ)_/¯
<ghane> haha ok
<ddevault> finished my part, s'all good
<ddevault> many thanks
<ghane> great, happy to help
<ddevault> ten points if you can guess what it's for https://l.sr.ht/1zGb.png
<ghane> not really from the model... a handle for turning something? (faucet related based on the filename)
<ddevault> yeah, it fits over the handle on my kitchen faucet to provide more leverage so this blasted dutch water pressure doesn't soak my kitchen every time I turn it on
<ghane> :D
<ghane> Have you very high pressure there?
<ddevault> yeah
<ddevault> below sea level after all
<ghane> ah true
castaway has quit [Ping timeout: 246 seconds]