Yeah there is something chemically, intrinsically wrong with what is happening because the plaster coming out of burnout is barely even holding together you can crumble it in your hand like dust. You can't even pick up a piece without another part falling off of it. There is a guy on YouTube who did kind of a hack job of this but he was having the same problem and solved it by adding Portland cement, the type n mortar that I happen to have a bag of is basically the same thing.
This molding already uses two full tubs of plaster of Paris at $6 a piece, and over half a roll of filament at $20 a roll, so ideally I would like to not increase the size of the mold if possible. Also the reason I'm doing small test parts until I can figure out the crumbling plaster issue.
I'm not so much worried about getting a full burnout (1000f+) because the pla soaked into the surface of the plaster has tons of room to vent, if I can get the burnout to happen as low as 3 or 400° and keep the plaster strong that would be fine with me, I just need to do more testing.