This is a post about failing at the same thing repeatedly, which is most of making things and almost none of what gets posted about making things. The mask in question took nine attempts across about six weeks. Attempts one through eight were all wrong for different reasons and I want to write those down before I forget them, because I will otherwise repeat at least three.
One to three: orientation
The first three failed because I printed the mask standing up, face forward, the way it sits on a shelf. That is the intuitive orientation and it is wrong for almost every mask, because it puts the entire curved surface at a shallow angle to the build plate and every layer line lands across the cheek where the light catches it.
Print it lying at an angle instead, tilted back so the surface you care about is as close to perpendicular to the layer direction as you can get. Supports go on the inside, where nobody looks and where sanding does not matter. That change alone removed maybe six hours of finishing per attempt.
Attempt three also warped, which I initially blamed on the tilt and which was actually the room being cold. Print temperature problems love to disguise themselves as design problems.
Four and five: splitting it wrong
The mask is bigger than my build plate, so it has to be split, and where you split matters enormously. My first split ran straight down the centre of the muzzle, which put a seam along the most visible ridge on the whole object and directly across the point where the two halves needed to be perfectly aligned.
The right answer is to split along a line the design already has — where a colour change happens, where a panel edge sits, behind the ear line. The seam becomes part of the shape instead of an interruption in it. On the sixth attempt I split it behind the cheek markings and the join essentially vanished under paint without any filling at all.
Add alignment pins to the split faces in the model. Two small pegs and matching holes turn a twenty-minute alignment fight into a five-second push-together.
Six to eight: finishing
Sanding a printed mask is a long, dull, dusty afternoon and there is no shortcut worth having. Start at 120 grit only on the layer lines, not across the whole surface, then 240, then 400, then a filler primer, then 600 wet. Skipping a grade shows up under gloss paint every time, and a fox mask that is not glossy is not doing its job.
Attempt seven I sanded through a wall. This is the specific hazard of a printed shell — the walls are two millimetres and enthusiastic sanding on a high spot goes straight through into daylight. Feel for heat with your fingers as you go; the surface warming up is the warning that you are removing material faster than you think.
Attempt eight I got a beautiful finish and then rushed the clear coat in a cold garage and got blooming across the entire face. Clear coat wants a warm, dry room. It does not negotiate on this.
Nine, and knowing when to stop
The ninth one is not perfect. There is a soft spot on the left cheek where the filler primer sits slightly proud, and I can see it in every photo. Nobody else has ever mentioned it, including people who have held the thing.
The temptation at that point is to go back to sanding, and the answer is almost always not to. A mask is worn on a moving person, at a distance, under bad hall lighting, usually for a few seconds in each person's field of view. There is a level of finish beyond which you are working for yourself alone, and it is a perfectly legitimate thing to do — but do it knowingly rather than because you cannot stop.
I stopped at nine because it was good and because I wanted to wear it, which turns out to have been the correct call. It has been worn four times now.
Nine attempts sounds like a disaster and was actually just a normal build, spread out enough that I noticed. Almost everything I know about printing came out of those six weeks, and the next mask took two attempts. That is the trade: the failures are the tuition.
All nine failures, plus the one that worked ⛩
VIP Access