Filament cooler fan - Can be speed up ?

I have installed a filament cooler fan with duct.
The fan I used is a common 60mm computer fan, three wires, +, - and signal.
It was connected to the PWM fan socket on my 2560 board (BTW the wiring diagram that comes with my geeetech I3 pro B is inverted so + goes up and signal goes down...).
It works fine and I can control the fan speed from repetier host, however very little air is pushed out of the duct, due to the fan turning too slow even when I set it at 100% on the Repetier Host.
Slowing makes it totally useless so I guess there must be something wrong somewhere. Possibly not the right fan ? Need edit the firmware ?
Possibly this depends on the fan being intended to be run at 12v instead of 5V - which I gather is the max output of the PWM socket ?
Anyway to improve the speed of the fan ?
What If I connect + and - to a 12v outlet somewhere and leave the signal wired to the PWM port ? Will I then obtain a 0 to 12v operating range instead of 0 to 5 ?
This is a common workaround in RC (leave the signal on the receiver and hook + and - straight to battery). Not sure this may work also with my printer, if I may damage something, or if there are more "stilish" ways to achieve same result.

The part cooling fan ought to be connected to the D9 terminal on the RAMPS board, and should be rated at 12 V. Power and GND connections only; leave the signal pin on the fan floating.

What do you mean by the PWM port? Did you connect it to one of the servo headers on the RAMPS board? Those only provide 5 V outputs. If you could attach a photo or schematic that might help. Don't connect any motors, heaters or even fans directly to the Arduino MEGA; you might damage its output drivers. All actuators should be hooked up to the RAMPS shield only, not directly to the Arduino.

If your fan has PWM signal pins in addition to the supply pins, then leave the PWM inputs unconnected--you're supposed to connect the + and - wires directly to the corresponding D9 terminals or to 12 V and GND rails, respectively. The D9 output voltage can be PWM controlled in G-code or via Repetier Host in order to adjust the fan speed.

EDIT: None of the above applies to the GT2560 control board for the Geetech Prusa i3. It may be of help for someone looking for info regarding a RAMPS setup, however.

Edited 6 time(s). Last edit at 02/23/2017 01:31PM by scndctr.

Here's the wiring diagram of my printer.

[www.geeetech.com]

The port marked as PWM-FAN is where the part cooling fan is to be connected - and where i did connect it.
I connected all the three wires.
I can control the fan from the repetier host manual interface - on off and % - and I guess no problems with the firmware or Gcode either as it starts right when the axis move on to start the print job, so it is recognized and acts as supposed to.
Only problem is it spins slower than it should IMHO, so not much useful.
I have NOT checked the voltage on the PWM-FAN port. I assume it is 5 volts as that's what I gathered looking around the internet for infos at to where the cooling fan was supposed to be hooked up, but I could be wrong.
I dont feel confident checking the voltage as the pins are quite close to each other and I fear shorting out.

Sorry, I read the "2560" in your OP and assumed you had an Arduino MEGA + RAMPS 1.4 setup, but this seems to be an integrated board with a completely different pin-out; my mistake.

After reviewing the schematic and the GT2560 wiki, it looks like you have the fan connected to the right port. It's possible that the signal line on your fan expects a different signal level than what the port is putting out; you'll have to look into the fan specs if you can find them. Or the issue could be something else entirely--sorry, I'm out of ideas.

Testing voltages normally shouldn't be a problem if you're using a DMM with pointy probes, you just need to be a bit careful not to have your hand slip and accidentally cause a short. As a matter of fact, you ought to be able to connect to any signal ground node on the board a safe distance away from the PWM port and still conduct a voltage measurement while minimising the risk of a short. Alternatively, if you have any spare connector parts and pins lying about, you could make up a jumper cable so it's easier to test those pins.

Edited 1 time(s). Last edit at 02/23/2017 01:28PM by scndctr.

It is not obvious whether you are talking about an extruder cooling fan or a parts cooling fan.

Most extruder cooling fans will run when the printer is powered up. Parts cooling fans are controlled by settings in the slicer. First check that you have cooling enabled and then check the layer at which the cooling cuts in. Most slicers will default to the fan being OFF for at least the first layer (which is correct). Therfore the fan will not run before the layer that it is set to start.

Edited 1 time(s). Last edit at 02/23/2017 07:14PM by Supermec.

i use this board with their rostock g2s printer...
among G+ users, 3 most common board problems are:
- burned power/bed connector (some of this has been fixed with terminal revision),
- blown bed mosFET,
- blown element "q5" (i can't remember if it's mosFET or voltage regular, but its providing power for PWM fan).
most cases of blown q5 are accidental shorts or overload do to multiple fans...

presonally most I've hooked up on it, was 80x80 6W delta fan on 12V setup.
at full power (100%) fan should get input voltage of the board, if jumper is set to 12V.
if the jumper is set to 24V, power for fans gets regulated down to 12V...

maybe you have duty cycle extra divided in FW?
you can easily check if you plug the fan in one of the other fan plugs... only one is PWM, others are constant on... so if it spins at same speed as PWM at 100% board isn't your problem.