seems like we had many same idears. I planned to do the same
modifications on the z axis and platform.
The partially closed air flow for the printhead looks interesting.
Would be nice if you can give me the scad file to print it.
I dont know if the modification on x and y axis are better.
After the same problems with noise and backlash (as you have described)
i found out that i only had to do a closer contact and (as a result
of it) more power to the stepper to fix the problem.
After it the noise was reduced to 50% (the vibration of
the backlash produces most of the noise )
If i would have a belt like you i would suggest that 2 bad new effects could happen:
1. a little backlash (material is not that solid as metall) + side effects (especially on higher print speed ) difficult to measure.
2. problems to place a noise reducer on the motor because your
motor has just contact to little teeth. (maybe thats why you have
removed the original one). Very interesting that you say that the
printer has lower noise now. Maybe lower than incorrect adjusting
but maybe not lower than my printer.
Actual my printer has not more noise than a normal 2d printer.
Maybe in a wire or belt-design you can move the x-motor to
the y-motor position and the y-motor to the chassis.
This would reduce weight extremly with lower problems.
sorry if my english is bad.