I'd be sticking it onto the back or side as you suggest but with a bit of double-sided tape so you can rip it off and move it if you want. I'd keep it off the extruder, as part of the point is to minimise the carriage mass, right?

I'd also leave the tube separate from the wiring, if you bundle everything together too neatly it will get less flexible, and it probably will want to follow a different path anyway. Experiment with locations, moving the head around the bed and see what seems to be freest / gets tangled up the least.

I've discovered that laser cutters use air-through-a-tube for what they call "air assist" (blows air out of the end to keep the burnt smokey stuff away from the lens), but usually the pump is more like a compressor than a blower. I think the pressure is necessary in this case though, while the cooling jet for the printer probably only needs a smallish flow rate...

Where do you think would be a good place to mount it? It's 75x75mm in size I was thinking at the bottom of the case and run the tube up the side and along the extruder wiring. But I'm worried the tube would be too long and I won't get enough air flow at the nozzle. Maybe on top of the heatsink on the extruder? Hmm but then it would bump the side carriages probably.

I have a plexiglass sides on the printer. Maybe on the side some how like superglue

The stepper drivers have thermal protection built in which shuts them down temporarily while the temperature goes back down - this presents in prints as layers offset (if x/y driver, common) or missing/partial layers (extruder driver). Keeping them cool prevents this from happening (and it's most likely to happen during a big print and waste a bunch of your time and plastic). The little stick-on heatsinks are a great idea (make sure they don't short anything out or prevent you from adjusting the vref trimpots), and if you're looking for a permanent and tidy solution for airflow I'd recommend putting a regular computer case fan (any old one will do) over the board (print yourself a mount for it) and wiring it directly into the 12V power supply wherever's convenient.

Regarding the blower and tube idea: blower and fan are electrically the same, so follow the usual routine to g-code control it (mosfet board if required, firmware, etc. - depends on your setup of course). Two-pin is fine, red=12V black=ground. 3-pin fans have a speed sensor on them that comes back through the yellow wire, even computers usually only use this to confirm it's running for protective reasons! Closed loop control certainly not required...

Blowers usually give better pressure than axial fans, so it's a good choice for what you're planning. I have no idea whether it's been tried before. The tube will be a fair bit of flow resistance, but of course it's worth the experiment. Make the tube as large a diameter as you think you can get away with (but then there's the weight/flexibility thing to tradeoff). Have fun, please share the experience!

Also was wondering if a small 12v 2-pin DC blower fan like this
http://www.ebay.com/itm/171266440063?ss … 1439.l2649

could be g-code controlled like a normal fan. If so, do you guys think I could print a "funnel" to put on the blowing end of it, and attach a tube to it and run the tube to the nozzle to act as an extruder fan?