Hi everyone. So I’m trying to 3d model a samurai helmet for 3d printing with multiple pieces that fit together. When I use subdivision surface modifier (even after turning down the intensity and playing with the levels) the integrity/shape of the holes for the other pieces to fit into no longer align. How would I go about smoothing the overall pieces without affecting the holes?
Welcome 
…
thsi depends on how you holes exactly are modeled… for example here:
the “edgy” part has “hard edges” (with edge grease ) to preserve the shape while the other areas are subdivided…
You may have to elaborate this to get a more suited answer. (And maybe take a little time here to be abelto post some images as a new user…)
Welcome! 
Have you tried adding edge creases to the holes and other parts that you want to keep their shape, or marking them sharp?
3d printing does not require Subdivision Surface Modifier to be maintained, so you can use Boolean after Apply. 
I don’t think it will be easy to make connecting parts during the modeling process.
And when using Subdivision Surface Modifier, it will be difficult to predict the contraction range that occurs in the connection.
There is a process of creating a connection in the middle of the video.
Here too, we use Boolean for objects with dense mesh structures that have been modeled.
In addition, thin-surface objects such as helmets do not make connections, but are glued to each other using glue after printing.
(Magnet is used for the front and back combination method for wearing.)
※ I’m not an expert in 3d printing, so I’m not sure if this is an appropriate process. 
Hi everyone, thank you for the responses. I have used the crease function and it helps to a certain extent. What I’m gathering is that I should close up the holes in the model that I created from the Booleans and Subdivide first. Basically, 3d Model->Solidify Modifier->Subdivision Surface->Booleans?
I applied the Boolean modifier before subdivision surface and you can see the result between the first two images.
I think you’re worrying about how things will connect and go together too early in the process.
The first thing you need to do is actually get the mesh sorted. Basically, however the mesh looks in viewport without smooth shading, is how the model will print.
Subsurf modelling requires clean topology. You can bodge it a little, but without a really solid and even underlying mesh, you’re always going to have problems. I can see from your screenshots that you’ve got lots of long rectangles, and no edge flow. Try and fix these things first.
I’ve done a lot of modelling for 3D printing, so I’m happy to help and offer suggestions, but you need the basic to be right first.
Thank you so much for responding. I admit, I’m new to this as this is my first attempt at 3d modeling. I do understand that it’s important to have quad topology, in the long rectangles that you see in my model, should I add more edge loops? Could you elaborate a little on what edge flow is?
This is what I have so far in my viewport as the whole model. I appreciate any help you can give me, I admit I’m trying to make sense of it all.
This looks pretty solid, tbf, just quite low poly. How it looks now is how it will look when you print it. This is why the subsurf modifier is important.
By edge flow, I just mean that the topology should follow the shape of the object (see my quick diagram). By doing that, you have more control over the overall shape when using subsurf.
Also, utilise bevels. Not the modifier, but actual bevels. Those hard creases you’ve got, if you select them (together, not individually) and do a bevel with 3 loops, and make it quite tight, it’s going to maintain those edges even with a subsurf. You might get the odd triangle face here and there, but ultimately, it doesn’t really matter as long as it looks ok visually.
Basically, go through each part individually, and get them looking nice and crisp. You don’t need millions and millions of verts, just enough to make the geometry less visible.
Thank you for saying so, I’ve been working on it for the past month. Sorry for the slow response, thank you for getting back to me. With the edge flow, I’m gathering that you’re just saying that I need more geometry to work with. Not just the edge but each piece just needs more faces overall? That being said, if I were to add edge loops to do that (not sure of any other way to do so) how would I maybe go about controlling the edge loops more accurately in those more difficult areas like curves and such? Is there a quicker way to add more geometry or just manually with edge loops?
I’m not as familiar with the bevel function just yet, I’ll have to look into that and try it out. Sort of like this?
Here’s a quick example regarding edge flow which might help. You can do what you’ve done (bevel the edges) but it’s perhaps a little cleaner (and easier to control) if you model it with flow in mind.
Here are two very similar meshes with two levels of subdiv:
Here they are in edit mode:
When we add thickness, look how much easier it is to control the sharpness of the mesh on the left:
A couple of edge loops, which, if I wanted to, I could easily remove by desolving the edge loops, keeping the original topology intact.
As for the one on the right, you could select the top faces, press I, and inset to get more or less the same thing, but it’s not quite as clean, and depending on underlying topology, it’s not necessarily going to be as easy to work with.
Comparing the two
At the end of the day, it’s entirely up to you, but hopefully this visually shows what I’ve been talking about in a way that makes sense. All in all, as long as it works, it doesn’t really matter
I think I understand what you’re saying, I’ll watch some youtube videos on edge flow as well to get a better understanding. I suppose having quad topology that more closely resembles actual squares would look cleaner and more uniform that what I’ve done with my model.
I hope this is allowed in this post, if not I understand. But another issue I have been having is when I apply my mirror modifier on any given piece and then pull up the 3d print toolbox and check for non-manifold edges (there frequently are some), click “Make Manifold”, it ends up making half the piece/model disappear. What would be causing this do you suppose?
Make sure you have clipping enabled on the mirror modifier. More than likely, it’s applying the mirror, but you’re being left with two individual meshes, rather than a single joined one.
Also, just my own opinion, I’ve made many, many things in Blender for 3D printing, and not once have I used the 3D print toolbox. I’m sure it has its uses, but it’s really not necessary.
I’ll make sure going forward that I have it turned on. In the event that I do and it still gives me that result, what could cause that do you suppose?
How do you personally check for non-manifold geometry, Intersect and zero faces?
Well, merge is on by default with mirror, so clipping only really ensures that the center verts stick, so it’s possible that you might still have some that have either over or under extended their middle line. Easiest fix would be to select the central loop (after applying mirror) and doing a merge by distance. That should fix the issue if it still crops up.
As for checking manifold, take what I’m about to say with a ginormous pinch of salt, as I’m just sharing my own experiences.
3D printing is actually a lot more forgiving than people think it is. People get very bent out of shape about so-called non-manifold geometry, but depending on what and how you’re printing, most of the time, it doesn’t matter. Obviously you don’t want holes in your meshes, so yes, the toolbox is probably great for that, but intersecting geometry also counts as non-manifold, and in regards to printing, is completely fine.
Is it best practice to make nice clean, watertight meshes? Sure. But if you’re just printing for yourself, then you only need to do what works.
I greatly appreciate all of your help. I admit I’m a little lost and if I’m being honest I’m probably getting in over my head with my first project. I’m attempting to design a 3d printable armor set from a late 80’s cartoon/anime for cosplay and maybe selling as stl files. Here’s some images of what I’m trying to build:
You seem to be doing okay so far, so I definitely wouldn’t say you’re in over your head at all.
Going back to my first post, model it first, and then worry about making it 3D printable. Utilise the Solidify Modifier, giving you better control over thickness (I’d recommend at-least 2-3mm thickness), and model in real world scale if you can (this is actually quite important for 3D printing). If you need a human base mesh to use as a fit guide, then the Blender Foundation have some free ones on their website you could use. (I’d recommend that for armour, removes a lot of guesswork).
Thank you for saying so. I just hand draw my own reference images with a ruler and a mechanical pencil and scan it and then I followed Tami Coker’s Iron Man modeling video on youtube and that’s kind of where I’m at. I realized my URL didn’t stick in the last post but this is what I’m trying to model:
As being a 6’3 tall guy, how would I go about scale?