E3Dv6 cannot keep temperature with fan on

Discussion in 'E3D-v6 and Lite6' started by Michel, Jan 5, 2015.

  1. Michel

    Michel Member

    Joined:
    Nov 7, 2014
    Messages:
    4
    Likes Received:
    0
    Hi all,

    I upgraded my Kossel mini with a E3Dv6. I am busy calibrating and while doing so I noticed I get severe undershoot of the temperature as soon as the PLA cooling fan switches on. Temperature then drops from 205C to ~195C and rises and falls in this range during print, messing up my print's finish in the process :( .

    On my E3Dv5 I used a 40W/24V heater, on the v6 I used the supplied 25W/24V heater. The fan's airflow does flow over the underside of the heating block a bit.

    I was wondering; do I need to PID auto tune with the fan on?

    I didn't have this issue with the 40W heater cartridge so I suppose I could go back to that one but I'd rather keep the 25W cartridge.
     
  2. Aggresive

    Aggresive Member

    Joined:
    Oct 10, 2014
    Messages:
    2
    Likes Received:
    0
    My fan just blows cool air over everything, which effects the hotend temperature a lot, so I set a PWM of 40%, when the fan kicks in the temperature drops 10 degrees (from 220 down to 210 - I run a high flowrate), but soon raises up again. However, if your fan is blowing too much cool air over the hotend, it will never get back up to temperature.

    Is the 'PLA cooling fan' PWM controlled by the firmware? I ask as the ranging issue that you get (205-195-205) would be caused by the fan slowing down and speeding up at different points in the layer.

    This could also be caused if you're varying flowrate (mm/s) throughout the layer.

    The 40W cartridge heater would have been able to put more energy/second (watts) into the heater block to compensate for the airflow of the fan.
     
  3. elmoret

    elmoret Administrator

    Joined:
    Mar 1, 2014
    Messages:
    586
    Likes Received:
    91
    You absolutely do need to retune the PID gains with the new setup.
     
  4. Michel

    Michel Member

    Joined:
    Nov 7, 2014
    Messages:
    4
    Likes Received:
    0
    I nailed it; I found my PID_MAX setting to be set to 190.
    After setting this to 255 and re-running the auto tuning the temperature remains at a rock-solid +/- 1 degree C regardless of fan speed (I use PWM but only to switch it on/off at 95%).

    Interestingly the fan doesn't change a lot PID-wise, so I kept it at the off values;
    Fan off:
    Kp: 47.14
    Ki: 6.60
    Kd: 84.18

    Fan on:
    Kp: 46.13
    Ki: 6.31
    Kd: 84.30
     
  5. digimatic

    digimatic Member

    Joined:
    Jan 5, 2015
    Messages:
    4
    Likes Received:
    0
    Are you talking about the supplied fan block which cools the hotend fins or the print cooling fan ?

    I'm running the supplied hotend fan on a separate supply to the standard print cooling fan which is on the controller, I have that one ducted away from the heater cartridge blowing down just below the nozzle. (I'll take a photo tonight) I think you really need to duct the print fan to get reasonable results (before I ducted mine, using the print fan usually generated worse results than no fan, with prints getting distorted on the fan side..or the bed getting overcooled and prints lifting)

    I am running the supplied fan for the 12v E3D on a 12v fixed feed as I think that needs to run all the time the hotend is hot right ? I'm only using a 12v/20W heater (not the supplied 40W one). Also might be worth checking you don't have a low heater PWM limit set in the configuration parameters which means the heater can't keep up ?
     
  6. digimatic

    digimatic Member

    Joined:
    Jan 5, 2015
    Messages:
    4
    Likes Received:
    0
    Ahh right that explains it :)
     
  7. Michel

    Michel Member

    Joined:
    Nov 7, 2014
    Messages:
    4
    Likes Received:
    0
    This issue was about an additional fan cooling the extruded PLA: the v6 setup is setup per the documentation.

    To be complete, here are the PWM-MAX 190 PID numbers (I mistakenly called PWM_MAX PID_MAX in above post);
    Kp: 39.74
    Ki: 5.23
    Kd: 75.54
     

Share This Page