Printable View
Hi,
Hopefully this is in the right folder!
I've just purchase a DaVinci for printing 3d conservatory and sunroom models.
I have never done this before and have a quick question - will the model below print ok or do you think the roof will fall when it begins to print because there is no structure underneath it?
http://3dprintboard.com/attachment.p...tid=4936&stc=1
Thank you in advance for your assistance.
Most FFF printers will print that fine without support.
I disagree, that roof looks like a fairly steep overhang and the window openings in the walls/doors are definitely going to be a problem. With active cooling it may be possible but you'd probably get best results adding supports. A finished part like that is open enough to allow for removal of the support and a light sanding to smooth everything out so it looks nice for your clients. Alternatively you could print it very small and rely on bridging but I imagine that would defeat the purpose of having a physical model.
If that's printed with 4" on each side, each window is less than an inch of bridging. The roof looks like about 35 degrees. I don't see a problem.
Post the .stl and let's see what we can do.
Unles you print too small and the various beam sections are under 5mm² it should be fine.
Alternatively : cut your model in 2 at the windows top Z. That way no overhangs for the windows and you can put some support for the roof.
Quote:
Originally Posted by Davo
HA the davinci won't. It won't even print it correctly without a roof. It doesn't like thin vertical objects like those studs.
First off, Lose the guy (print that separately if you want), run the shed model through solid inspector (a plugin for sketchup), fix any of those errors (I can see a couple already), and then run it again.
It also appears that you bottom wall has 0 thickness. While it will print, it will have zero strength and almost instantly delaminate.
The model is excellent as a virtual model, it's beautiful. But there's some issues with printability, ESPECIALLY with the davinci. (I own one, I both love it and hate it). If you're going to print it on the Davinci, you're going to have to flip it over, onto it's roof, and print with supports. Otherwise you'll have to spend days gently sanding off support material, as your design is not structurally strong in ABS (ABS issue, not your design), and it will come out one big block of supports. There is no way (that I know of) to selectively print support material on the Davinci. Anything beyond like a 10 degree overhang gets support. So the entire interior and every window frame will be supports. Flip it, and it will use less support material.
Post up the Sketchup file, and I'll run it through Solid inspector to see what the damage is. It may be bad enough that you might want to start from scratch with printability in mind.
I would definitely recommend downloading or designing progressively harder to print models. Learn the capabilities and limitations of your machine before starting something complex.
Good morning,
Thank you to all the replies. I thought I would share an update;
1. I removed the roof and sliced the image in Da Vinci's own software - little did I know it didn't work and simply tried to print a cube!
As soon as I realised that it was just printing a cube upwards I stopped the printing and did a little research - it seems XYZ's own software isn't well liked amongst budding 3d printers!
2. I bought Simplify3D
Their slicing tool and ease of use meant I could understand the printing technique a bit better.
3. I removed the roof and overhangs
As I type this I've clicked print and will post an update on how it went!
Thanks, please keep us posted. Your findings will help others in the future.
:)
I will be VERY surprised if you get a 3rd party slicer to work on the davinci, as they use a proprietary version of gcode that isn't public as far as I know. And if you are using a 3rd party slicer, I would firewall both the XYZware, the printer, and anything else they installed. They have been known to remotely brick the printer if 3rd party apps and hard ware are detected.
Did you try any of the suggestions I made previously? I can see multiple manifold issues with your design from the pic alone, Solid inspector will identify them from within Sketchup.
Yeah sort of what Marm said - sketchup is really crap for 3d printing models.
It's just not a3d solid modelling bit of software and all printable models need to be solids.
I'd like an stl of that original model though just to see if I can print it as is.
Should be able to. If it's a proper solid and not simply a collection of faces.
I have to disagree on that CA.
BUT... it's all I've used for 3d printing so far. I know it's not the best for it, but I do ok with it. I've had to learn how to make it work for printing. Maybe when I move onto another program I'll realize what you mean, but I see nothing "wrong" with it, 3d printing just isn't it's forte. I can whip up something really quickly in SU, convert it and print quickly now. Just gotta think like a 3d printer from the get go. Converting old designs to be printed..... that's another story.
Why not print each wall and roof section flat and glue together post print?
Kind of defeats the purpose of a 3d printer then IMO.Quote:
Originally Posted by jlmccuan
The other problem I see is the lower rail caps (where the studs base onto). Its completely horizontal. A support fill will be needed there. It can be removed later after the print is done though. And what about the top of the windows, thats horizontal as well.
Well, you build houses that way and they are certainly 3d. The walls would still be constructed of 3d printed framing. And printing that way would give the best quality end product. Just because you might be able to print something in one pass oriented in it's final direction doesn't mean you've delivered the best product you can.Quote:
Originally Posted by Marm
I guess if you can get sketchup to do solids and you can make it work (I never could, none of the tutorials worked) then it should be good.
I'd still like to try and print it. :-)
There are a couple pitfalls you have to look out for.
One is hidden interior geometry. If you end up with lines inside a solid, it will confuse a slicer (at least mine). Apparently, a single line in SU has enough thickness that a slicer is trying to treat it like a cylinder of the minimal diameter. Imagine connecting the centers of opposite faces of a cube with a line. The SU and the slicer try to reconcile this as a solid object, that is part of the manifold object. Same with buried faces. If you end up with two intersecting faces, but the intersection endpoints aren't connected by a line, then it will create one or more internal faces that again confuse the slicer. By simply drawing a line between the intersecting end points, and then deleting the resulting isolated internal faces, you will avoid this problem.
Another is issue is masked faces. Imagine making a cube, then making a bigger cube, then you take the smaller cube and move it share a corner and a plane. The two faces that share that plane are causing an issue with the slicer. SU still treats each cube as a solid object, with all outside faces, but they have 0 seperation. The slicer tries to do the same, but with overlapping faces like this, it's the equivalent of dividing by zero. The solution is far easier than interior faces/lines, as you just have to delete the overlapping faces, leaving the interior space contiguous.
Redundant lines don't seem to be as much of an issue, but I still like to remove them. If you have the same scenario as above, but with cubes of the same size, you'll be left with a square of lines that wrap around the midpoint of the new polyhedron. They don't seem to do anything, nor confuse the slicer, but I still take them out for a cleaner appearance.