Edit 2013/08/26 23:25 EDT: I just realized something: what I've said below about vibrations and steppers still stands; however, since our machines are driven by belts instead of something rigid like screws or gears, the steppers are much less likely to be affected by vibrations. The belts will have an attenuating effect. But, I have read about some users experiencing vibration-induced missed steps, so I think that if your fan produces *any* noticeable shaking when powered, do yourself a favor and put a dampening gasket in there. That goes for anything that you add to a moving part of these machines. As it usually goes...better safe than really pissed off later!
Nylocke wrote:When you say "balanced", are you referring to the bearing medium, or something else?
I suppose it would have to do with the bearing. I'm just wanting to note, for those that are unaware, that vibrations *can* cause hell for stepper motors. When the rotor moves between step points it acts under the principles of a dynamic system. There are tiny oscillations around the stopping point that can be amplified if affected just right. Since the holding torque when micro-stepping is less it takes a smaller "kick" to miss a step.
So if your fans noticeably shake when powered, I'd add some kind of dampener between the fan and the mounting surface in a matter similar to adding cork between your steppers and the machine frame. You might also notice a "kicking" of the fan at low speeds (low PWM duty cycles), which could also be an unintended source of vibration, even on a normally well balanced fan. I'm pretty sure the "kicking" is a side effect of a low PWM frequency. That is to say, with the same duty cycle and a higher frequency, you can get a "smoother" output of the same voltage.
Now I'm curious: I wonder if the AVR's hardware PWM outputs are being used for the fans and heating elements or whether they are doing some kind of software PWM generation? To the code batman!
@flateric: I hadn't thought of that directly, but I like the idea. This duct is, I think, v5.2

The other revisions had different nozzle shapes, but mostly due to my inexperience with surface modeling in ProE. Most of my CAD work has been very manual-machine-shop oriented, i.e., right angles, some fillets that can be done with ball nose end mills, circles, squares, straight-lines, things that can be made on standard manual machines. I had to dig out the old texts to refresh my memory on sweeps, blends, and b-splines, and as I regained proficiency, the nozzle evolved.
I can comment on a fine tipped nozzle, though, from my next-gen (what I'm using now) cooling trials. Smaller orifices require more pressure to maintain sufficient flow, and fans are terrible at generating pressure. Not saying it couldn't work, but a couple of tries with something similar didn't.
I did have an idea once about cooling with an aquarium pump. I'm not sure if you could switch it with a solid-state relay to get the regulating effect. I'm also not sure if that would be bad for a pump like that. Not so much the motor, but often small pumps need to be pumping to maintain adequate flow for cooling.
I won't promise anything as the purpose of getting my printer was to aid prototyping at work, and I seem to be perpetually behind due to self-inflicted wounds. But, modifying computer files is rather trivial...I just need to think up a quick-connect feature for changing nozzles. Maybe I can create an easy-to-duplicate mating surface so that other people could develop their own nozzles with the software of their choice...