Prints pulling up at corners

Need some assistance getting started...post here...
naturalstate720
Posts: 77
Joined: Thu Nov 05, 2015 7:57 pm

Re: Prints pulling up at corners

Post by naturalstate720 » Tue Nov 17, 2015 7:34 pm

That makes sense. I wish that was explained to me at the beginning of this frustrating process. Is there any kind of filter or something I can apply to the model to print it exaggerated in one direction so that when it shrinks it shrinks perfectly square?

User avatar
Jules
Posts: 3144
Joined: Wed Jan 21, 2015 1:36 am

Re: Prints pulling up at corners

Post by Jules » Tue Nov 17, 2015 7:41 pm

Is that part of the building hollow at the base?

How thick are the walls at the base where it contacts the bed? Are they really thin? (Just a few widths thick? Same thickness as at the top and your window wells?)

If so, it's no wonder that keeps warping, and i'm amazed that you've gotten prints as good as those are. (That last one looks pretty good, all things considered.)

If those are thin walls at the base, thicken them by at least 10 mm. Even better would be to make them solid. Just at the base, and you can put the thick part inside, so it's not visible. Make them about 4 mm tall for good stability.

If those are as thin as they look, there is literally nothing holding that down to the plate except the brim, and that's not enough to hold down a structure that high against the "forces of warp" that ed and jim mentioned. You need enough solid surface contacting the plate to keep things stable. I thought the base of that was solid, like the bottom part of the building, but that picture makes it look like it's hollow.

Maybe it's just the angle of that last photo? :?

If it's not proprietary, and you wouldn't mind posting the STL file (you'd have to Zip it up first) I'd like to take a look at it. It should be possible to attach a thick band to the inside in Rhino, although no promises.

User avatar
ednisley
Posts: 1188
Joined: Fri Apr 11, 2014 5:34 pm
Location: Halfway up the Hudson
Contact:

Re: Prints pulling up at corners

Post by ednisley » Tue Nov 17, 2015 10:01 pm

naturalstate720 wrote:so that when it shrinks it shrinks perfectly square?
That's what I meant when I said you must adapt your model to the process. The model is the only place where knowledge of the overall design and the construction process come together. The slicer gets a file full of triangles and some overall speed-and-temperature guidelines, so expecting it to bend your design to null out the stresses left over after you peel the plastic off the platform won't work: the slicer can't know enough.

The shrinkage will be roughly zero at the platform surface (it's glued in place), essentially zero vertically along the Z-axis (because the nozzle is always at the right height), and locally increasing across XY as Z increases (because the upper layers are less constrained). However, the part won't shrink globally in XY, because the lower layers hold the upper ones more-or-less in place as the plastic shrinks locally: that's what freezes stress into the finished part.

If you insist on printing huge parts in one shot, then you must make the joint tolerances large enough to accommodate the inevitable distortions. The walls probably look pretty good, so if you make them fit together, the final assembly will be acceptable. Heck, nobody expects girders to fit with micron precision; they drill those bolt holes slightly oversize for a good reason.

Architects used to work with razors, foamcore, and clay, so there's a definite crisis of rising expectations going on here... [grin]

naturalstate720
Posts: 77
Joined: Thu Nov 05, 2015 7:57 pm

Re: Prints pulling up at corners

Post by naturalstate720 » Tue Nov 17, 2015 10:44 pm

Most of the building is completely hollow. Half of the bottom is a partially solid rectangular box with heavy infill. The other half is a complete car entry and exit for robotic underground parking garage. It's a void space that spans the length of building. Boss really wanted to show it like that, so there is only a thin wall on one side.

The feedback makes sense. Are there any good articles about all this stuff? Best way to prepare model, geometry, breaking it into pieces, and trying to reduce warping?

naturalstate720
Posts: 77
Joined: Thu Nov 05, 2015 7:57 pm

Re: Prints pulling up at corners

Post by naturalstate720 » Tue Nov 17, 2015 10:45 pm

Also, there was support generated throughout this car tunnel.

User avatar
Jules
Posts: 3144
Joined: Wed Jan 21, 2015 1:36 am

Re: Prints pulling up at corners

Post by Jules » Tue Nov 17, 2015 11:26 pm

So you have a garage underneath it that runs the length of the building, and you basically have one or two thin walls that go down to actually contact the plate on the outside? (Am i reading that correctly?)

And those thin walls are where the warping is occurring? (Sometimes?)

The support is going to help everything above it, it's not going to do a thing for something that is just effectively standing next to it, unless you actually attach the support to the walls. Then it might help to hold it down. Or it might not, support tends to be thin by nature in order to allow it to be removed.

I'm sorry, because if i had seen what you were trying to do sooner, I might have had some ideas that might have saved you a couple of weeks of lost work. But none of the photos up till now showed that hollow spot.

I don't know of any quick guides that would tell you how to design for 3D printing - the one thing that's helpful to remember is that the more surface area that you have contacting the plate, the less trouble you are going to have with prints pulling up when they warp.

Couple of things that might work:

1. Put a bottom on the garage. Have a solid surface that goes all the way under the building. And it needs to be merged with the design, attaching to both the outside walls and anything in the interior. Make it about 2 or 3 mm thick, and use about 4 or 5 bottom solid layers.

2. Thicken that outside bottom wall so that more surface area attaches to the plate. And by thicken it, I mean it needs to be about 10 mm wide to give enough surface area to hold it down. It's so thin, compared to the solid area, that that is where the warp is going to happen.
walls.jpg
3. Cut the design up into several pieces so that they are not so tall, then glue them together like the guys suggested. You will still want to put a lip or something on the inside of the parts so that the thin walls are securely anchored, and so you will have something to glue together.

This was not something you would have known to design for, or to ask about. In the future, you'll know to incorporate it into your designs. But it is something that people who create files for 3D printing work into their designs.

Update: That first diagram was rather lousy - i'd actually design something that looked more like this, so you could cut it off more easily after the print.
walls2.jpg

naturalstate720
Posts: 77
Joined: Thu Nov 05, 2015 7:57 pm

Re: Prints pulling up at corners

Post by naturalstate720 » Wed Nov 18, 2015 5:14 am

Well I was having legitimate warping problems with other models before this building, but I overcame those issues with the help from everyone. I printed an entire block of buildings adjacent to the project. I was having problems with warping on these too and they were just boxes with floors. Guess it was a combination of adhesion, zstop and bed leveling.

I realize I probably should have added a floor to the very bottom of building, that probably would of helped.

Luckily the model turned out well. The second part of the building wasn't warped (even though it as hollow without bottom) kind of werid. I got it glued on, and you can barely even see a line! Boss is happy! Now that I'm not in a time crunch on a project I'm going to go back and try to learn all this stuff right.

Thanks for the tips.

Post Reply