Round 2 of my estimations for the high efficiency boiler goal.
Here is a picture of the cylinder and cone stack pulled directly from the Solidworks field drawing.
I would guess that there is some sort of indexer at the top similar to the bottom feeder, so lets assume that only the cylinder and the cone can hold excess fuel, and the of the innards are used to organize and count fuel.
Boiler Stack Cylinder
Radius: 7.5"
Height: 30"
Volume: 5,301 cubic inches
Boiler Stack Truncated cone
Height: 9"
Upper radius: 10.875"
Lower radius: 7.5"
Volume: 2,413.5 cubic inches
Total Volume: 7714.5 cubic inches
Fuel volume: 65.45 cubic inches
Random close pack of a sphere:
~64%
Total capacity of boiler stack: 75.4 fuel.
The boiler processes fuel in the high efficiency goal at a rate of ~ 5 fuel per second. Which means it would take ~15 seconds to process it all.
Firing 8 fuel a second (A feat I see very few teams accomplishing), a robot would need 25 seconds to fill the high efficiency boiler, firing a total of 200 fuel with 100% accuracy.
The max volume of a robot (34,560 cube in.) would hold 338 fuel, which obviously is not feasible. 200 fuel packed randomly would fill ~ 20,450 cubic inches. This would leave ~ 40% of the total robot volume available for drive train, and shooter.