Sunday, September 29, 2013

Active cooling of J-Head insulator...

Construction of my self-build Mendel90 continues. Recently my new J-Head hot-end arrived, the final part of the jigsaw! Experience from my other printer has taught me that active cooling beneath the x-carriage has a number of positive benefits. Placement of a small fan that directs cooler air horizontally beneath the x-carriage causes a disruption to convected heat rising from the hot-end, preventing a temperature rise in the x-carriage that might cause deformation if overheated. It also cools the PEEK insulator on the J-Head which then maintains a short thermal transition zone, essential to prevent jamming. (There is a good illustration and explanation of thermal transition zone here on Nophead's blog.)

The current (at time of posting) Mendel90 x-carriage design didn't have a mounting point for such a fan. I had been designing a clip-on fan bracket when I saw Goopy's x-carriage modification to take a direct drive 1.75mm extruder, which included an under-carriage 40mm fan mount. Goopy's x-carriage wasn't suitable for me as he had also altered the carriage opening and extruder mounting-hole positions. So I imported the original x-carriage stl into Sketchup and modified it to meet my fan mounting needs. I also included M3 nut-traps into the design, which makes fitting the fan much easier, especially after the carriage is fitted to the printer. Goopy's lower shroud (half ducting) was of use as it directed the air flow above the stock cooling duct.
Illustrated above is the design change to the M90 x-carriage. 40mm fan mount. Design shared on Thingiverse.
Photo above/below shows first fit of 40mm fan and half-ducting.

Photo below shows new J-Head V in position. 

Below is a view from beneath the x-carriage, showing the new fan in position (right of pict) and the existing larger fan ducting for cooling the work. The half ducting on the new fan redirects air nicely up and over the larger ducting.
Finally below, a side view of the new fan mounted in position. It sits flush under the x-carriage resulting in no change in the length of the the carriage.

Controlling the fan:
The fan takes it's power via the ribbon cable to the x-carriage. The Mendel90 design has spare capacity on the 20 strands of ribbon cable that run to the x-carriage. I have the fan turning on/off automatically from the the Marlin firmware. A pin on the Azteeg X3 controller board becomes active when the temperature of the hot-end rises above 50deg C. This is all set is the ConfigurationAdv.h tab of Marlin (Section shown below.)

// Extruder cooling fans
// Configure fan pin outputs to automatically turn on/off when the associated
// extruder temperature is above/below EXTRUDER_AUTO_FAN_TEMPERATURE.
// Multiple extruders can be assigned to the same pin in which case 
// the fan will turn on when any selected extruder is above the threshold.
#define EXTRUDER_0_AUTO_FAN_PIN   5 //IOS 20130914 //-1
#define EXTRUDER_1_AUTO_FAN_PIN   -1
#define EXTRUDER_2_AUTO_FAN_PIN   -1
#define EXTRUDER_AUTO_FAN_SPEED   255  // == full speed 

The benefit of this is the fan operates automatically under firmware control, and I don't need to add any start/end g-code to drive it. It remains on after the job has finished, but automatically shuts off once the nozzle temperature has dropped below 50, or some other temperature specified in firmware section above.

Thanks for viewing!
Comments and questions welcome.


  1. Thank you for the post, I ran into the problems which you describe and fitted a 25mm fan which I clipped into place on the X carriage. My fan is controlled via an M42 command to toggle one of the Panelolou2 transistors. One (pedantic) point ... when you say "...disruption to conductive heat rising..." shouldn't this be "convected heat rising"?
    Neat work and a far better solution than a simple fan clip!


  2. Hi Alan,
    Thanks for the feedback, and picking up on the "conductive/convected" mix-up. You are correct. :-)