Gondola print, why are the walls falling inwards?

[Stephane]

Hello all!

My latest project was to print a gondola with fiberglass cover...

[ Guests cannot view attachments ]

Pretty excited because this is the very first 3D project where the model is essentially created by myself (the only part not built up by me being the brake wheel).

I knew the first print would likely fail, where I'd screw up the tolerances on the cover, or the trucks, but I figured I'd get a print done, and then fix the model for subsequent prints.  Actually, the cover fits perfectly ( ), and the trucks' coupler box doesn't ( )

However, there is one bit that I really didn't expect to fail, and fail really badly - the walls!

[ Guests cannot view attachments ]

The dimensions of the gondola walls are perfect at the top and at the bottom of the gondola.  However, as seen in the above picture, the walls got 'sucked' inwards right above the floor of the gondola!  I don't know what is causing this, or how to fix it, and I'm hoping maybe someone on the forum has seen this problem before. 

Note that the top of the gondola was towards the build plate (i.e. the supports were all along the top edge of the gondola, and some light supports attached to the floor of the gondola.  I figured that this would prevent any cavities and suction.. i.e.:

[ Guests cannot view attachments ]

Also note, the walls of the gondola are 0.5mm thick.. is this too thin?

any help would be appreciated!

[Mark W]

@Stephane, I've been pondering this one.  You have it positioned the way I would go.  There will be some strange forces right when those first few layers of the floor are exposed.  At that point, you've essentially made a bucket to scoop resin out of the vat with each layer reset. 

The only idea I have is to try adding a few cross beams between the inside of the gondola walls  Maybe .8mm x .8mm, with 1mm gap between the gondola floor. 
These beams would be cut away post cure, but the real test is to see if they help hold the gondola sides in their proper spot when it comes time for the floor to be exposed. 

[wcfn100]

I don't think that's the best orientation.  When the floor starts printing, you're basically creating a tub of resin that goes up and down with the build plate. 

I had one project that did something similar which made me starting thinking about how the resin would have to flow around the objects as it cycled up and down.

Try tilting the gondola at an angle (one end higher than the other).

Jason

[Stephane]

Thanks for the advice, turns out I got to much the same place in the end!

After posting the question, I started tinkering and trying version 2.  I increased the wall and floor thickness to a full 1mm, and then added a single row of 1mm diameter cross braces to the inside, roughly halfway up the side walls.  I also decided to print entirely flat, just to see if it would work.

It did print, but the single row of cross braces, while it improved the walls, did not entirely eliminate the warping.  Printing flat also introduced another problem, where the floor seemed to swell somewhat, and the ribs near the ends of the car were warping outwards.  I didn't expect that at all!  Look at the bottom of the ribs on the car, they are warping outwards the further near the ends.  The middle ribs (not seen in the picture) are perfectly aligned!  And this only happens when printing flat, completely disappears when printing at even 3 degree angle.

[ Guests cannot view attachments ]



So back to the drawing board (or, well, tinkercad).  I reduced the wall and floor thickness down to .75mm after finding 1mm too thick, and this time I went full lattice bracing!

[ Guests cannot view attachments ]

This is basically 0.6mm thick square profile bracing, 0.8mm apart.  They do not touch the actual floor of the gondola (actually, I find that it prints MUCH nicer if I don't support the floor at all!)

I then printed the same way (upside down) with 5 degree incline.  Note that I didn't want to print right-side up because of the suction effect of that large cavity of the gondola would produce.  Though I did have two drain holes - 2.1mm diameter holes for the trucks, I have them going right through the floor of the gondola 

It printed very nice this time around (finished printing last night, and I cleaned up the print late this afternoon).  The only problem I had is I didn't put enough supports, and the top of the walls sagged between the supports.  But after sanding down the top of the gondola against a sheet of sandpaper, this is barely noticeable, and easily corrected in my next print by doubling the number of supports around the edges!

[ Guests cannot view attachments ]

[ Guests cannot view attachments ]

I installed some micro trains trucks, but I have to change them, turns out I need the short coupler, not the medium.

This is what I was trying to reproduce - for copper/zinc concentrate transport:  http://www.rrpicturearchives.net/showPicture.aspx?id=4322869

Though please note, I am not really into perfect reproduction of a prototype, I'm happy with "close enough" 

Anyway, thanks for the comments, they really are appreciated!

[C855B]

This is a very nice result, and thank you for the step-by-step problem-solving. (As if I don't already have a ton of stuff on my platter,) your achievement is making me think long and hard about a Photon in my workshop for a couple of prototypes I have in mind.

[Stephane]

@C855B, thank you, you really do have to think long and hard, because if you get one, progress on everything else will suffer immensely    3D printing is a hobby by itself!