Well, almost. I didn’t give myself enough tolerance in the cutout for the speaker and it doesn’t fit well. On to v1.01!
Are you trying to build a replica of hoover dam ?
I don’t use tree supports much so I was hyper focused on the tree…
Oddly branches tree. Doesn’t look organic. Not a tree? Is it like one of those finger hand things? With tiny hands at the top?
Hands look like they’re holding up a- oh shit they’re tree supports!
PSA: stop doom scrolling right when you wake up.
I thought it was a homage to a puzzle from The Witness, hence the apple “for scale”. Even now that I know it isnt I still cant unthink it.
Edit : just scrolled down and somebody else did the same thing and also included a screenshot.
Sometimes the supports themselves just look so cool
He was printing a tree for his model train set. No idea what the wall behind it is for
That’s supports
With big prints like this that require tolerances I tend to make a dedicated model for just the feature with the tolerance and hone it in on that. Since just printing the bit I need to test is way faster than reprinting the whole thing a bunch of times. It takes a bit more planning but is totally worth it in saved plastic and time!
If I’m in a time crunch for the piece then sandpaper works well enough if the tolerance was too tight, and some wraps of electrical tape if it’s too loose
The print looks great tho!
I agree. In fact, that’s what I tend to do - slice up a design by splitting the body/bodies and printing test pieces where tolerances matter. Things like latches, hinges, pieces that have to fit with one another, etc. I’m not sure how practical this approach would have been for this print due to its final orientation, but it’s a really good practice.
I think I got a bit too comfortable with things going per plan over my last batch of designs :( I’ll also admit to being in a bit of a time crunch. No deadline, but I have younger kids so time to model and print is somewhat limited. This is a good reminder that rushing can actually make things take longer in the end.
Massaging this print to fit wasn’t practical. Despite being off by 1% that’s still 2mm of material to remove over some pretty big spans. I did take a chisel to the cutout, but man is ASA tough. PETG is much easier to do that with lol.
Thanks re: print looks great. It’s super solid, so I’m very happy in that regard. I don’t know about you, but lighting greatly impacts how the surface quality of my prints look. Hard/direct light at a steep vertical angle makes the faces look pretty rough, but more diffuse light coming from the side makes the parts look great. I am not sure if this is normal, especially for a larger CoreXY with long 6mm wide a/b belts, or if this is something I can dig into and improve.
The apple doesn’t fall far from the tree supports.
What is it?
It’s a stand to somewhat elevate and angle the speakers I used with my computer.
I’ll follow up with another post in a day or two with the finished product.
That’s what a small set of files is for! You put in three boundary layers, use them!
I put this in another reply, but I know not everyone will pop back into the thread so…
I completely agree with your approach and that’s what I would usually do. The print is probably off by 1%, which over these spans is 2mm. Massaging this print to fit isn’t really practical :(
no broken branches, whats the solution
In my experience, broken tree branches come from:
- Crazy angles on the supports. This happens when a support needs to ‘grow’ over the print in order reach the thing it needs to support. This ultimately comes down to part geometry, so there’s not a ton you can do here if you can’t change the geometry or orientation of the part
- Poor bed adhesion causing the trunk to separate from the bed. Clean your bed with dish soap and dry it with paper towels. Make sure you have a good first layer by getting your your bed and gantry in parallel planes and double check your z-offset. Bonus points if you can do a bed mesh between prints
- The extruder catching on a branch and breaking it off. This is usually due warping or over-extrusion. Warping can be its own rabbit hole. Over-extrusion is easy to tune for, especially if your slicer has built in calibration aids (eg OrcaSlicer, SuperSlicer, etc)
- An ambitious slicer not making the supports themselves very strong. Slicers these days seem to avoid thin/tall trees, but they’re still usually single perimeter. I’ve configured my slicer to use 0.6mm thick walls on supports
- If you have a bed slinger, tall supports can wobble. Slowing down acceleration/jerk is really the only way to combat this
Obviously, these can all be a bit interrelated.
The support in this print is basically vertical (no crazy angles), I generally have great bed adhesion/my printer can mechanically make its gantry in plane with the bed/I run a bed mesh every print/I use klipper_z_calibration to get a consistent first layer, nothing’s warping and I’ve tuned my extrusion multiplier for this spool of filament, the support itself is strong due to its girth at the base and wall thickness, and CoreXY means that the support doesn’t really move unless the extruder is dragging some.
Thank you for the very in depth response, I must confess I have zero knowledge in 3d printing and my comment was actually a joke about The Witness. Iit’s a puzzle heavy video game, and one of those puzzles has you figuring the correct pattern on a screen, by noticing apple tree branches just like the one you printed
It looks like this lol

lol, I see. Printer tuning is a very real struggle for some and it happens that tree supports are one of the things that you can run into.
That sounds really cool! I wouldn’t expect this kind of shape to be ideal for balancing
If it almost fits, can I suggest a bit of heat from a heat gun or even a hairdryer might soften it enough to allow it to fit
Sadly, almost is relative. The dimensions were off by say 1%, but over larger spans that’s 1-2mm.
I am not at all opposed to taking heat and/or tools to prints to massage them vs tossing the first go and printing fresh, but that wasn’t practical here :(
Bummer. Glad you had already considered it though. At one point in my 3D printing journey I would’ve gone straight back to CAD for small tolerance issues




