I've got the same printer - it's great, but clearly not optimized for power consumption. The hot end could be insulated so the only loss of heat is via the melted filament - as it is, I think most of the energy that goes into the hot end is lost directly to the air. Then the heated bed is great for ease of use, but it's pretty wasteful. If power was at a premium, I'm pretty sure we'd go for other solutions to keeping the print stuck down. My making both these changes, I'd guess it would be fairly easy to reduce the power consumption by a factor of ten or so. In the end, it ought to be possible for printing to take negligible power compared to making the bioplastic filament in the first place.

You don't need a heated bed. Just some glue stick or hairspray on the build plate

>by plastic mass, FDM is probably way less efficient energy wise than injection molding for the same mass of plastic

How does it compare when you consider the energy required to manufacture small-run injection molds?

Clearly at a certain scale there must be a crossover point, and it's probably greater than 1 unit.