my actual setup: Bambu Lab P1S + AMS
Most of these notes come from printing on a Bambu Lab P1S with the original AMS. I like the P1S because it removes a lot of printer-tinkering from normal jobs: the enclosed CoreXY machine, automatic calibration and Bambu Studio workflow let me spend more time on the part itself.
The AMS is useful even when I am not doing a multi-color print. I can keep several materials or colors ready and switch between them without manually unloading a spool every time. Because SUNLU is third-party filament, I set the material/profile manually in Bambu Studio instead of relying on Bambu RFID identification.
start with the first layer
Most failed prints are easier to understand when I stop looking at the whole model and look at the first layer first. A clean bed, the correct plate/profile and a consistent first layer solve more problems than randomly changing ten slicer settings at once.
mostly SUNLU: PLA first, PETG when the part needs it
I mostly use SUNLU because it is usually easy to find at sensible prices and has been a good fit for my everyday prints. I normally check szpulomat.pl when I want to compare current Polish offers and price per kilogram instead of sticking to one store.
PLA is still my default for prototypes, stands and parts that live indoors. PETG is what I reach for when a functional object needs more toughness, water resistance or a little more temperature headroom. I treat manufacturer profiles as a starting point and save a custom Bambu Studio filament profile when a particular SUNLU spool needs its own tuning.
orientation beats brute-force infill
Layer lines create direction. If a clip, hook or bracket will be pulled across the weak direction, adding more infill may not fix the actual design problem. I try to orient functional parts so the main loads work with the walls and layers, then add wall thickness before blindly pushing infill upward.
tolerances are part of the model
A CAD dimension that is mathematically perfect is not automatically a printable fit. Holes, sliding parts and press fits need clearance. For me, the useful workflow is measurement → small test piece → adjust → print the real part. It wastes less filament than discovering a bad fit after a multi-hour print.
supports are a design signal
Supports are useful, but if a simple functional model needs them everywhere I first ask whether the geometry or orientation can be improved. Chamfers, split parts and smarter print orientation often give cleaner surfaces and faster prints.
my basic workflow
idea ↓ measure the real object ↓ simple CAD model ↓ check orientation + overhangs ↓ slice + preview critical layers ↓ watch the first layer ↓ test fit ↓ revise the model, not just the slicer