A very low fidelity model came together over 12 hours. The general idea is the thrust structure is a bowl which collects the ashes. It has to be turned over to empty the ashes. Magnets attach the upper stages to the thrust structure. The upper stages are taken off to light the incense & put on after the flames go out. Then, they theoretically capture the ashes while allowing smoke to rise. The thing which collects the ashes has to be separate from the thing which is put on after lighting.
On the 1st try, the bed leveler immediately failed. It had been consistently measuring the front left corner .2 mm low for several weeks. Then, it inexplicably returned 200mm as the bed height & started printing in the air. It was disabled & the lion kingdom went back to manual bed leveling.
After another 2 hours of printing,
The extruder suddenly failed partway through it. After a day of troubleshooting, the best strategy was to manually press filament into the nozzle alone. Then manually press filament into the bowden tube. Then manually press filament into the extruder. Each step was incrementally harder to press filament into, but who knows how hard it was originally. The best results happened after replacing the bowden tube. It extruded a few layers before stalling again. The starship incense project was dead.
There was a lot of grease on the outside of the bowden tube from an unknown source. Grease could have also gotten inside the bowden tube. There, it could have added enough friction to stop filament from sliding along. PTFE is specifically used because it's slippery. The EZR struder has also worn down from abrasion.
Dehydrating PLA+ in the solar filament drier could have leached out oil. It could have made it harder to push through the bowden tube. It's believed a direct extruder has the best chance of solving the problem.