Chimera with active cold-end cooling?

Discussion in 'Multi-Extrusion' started by florian, Nov 23, 2015.

  1. florian

    florian Member

    Joined:
    Apr 27, 2014
    Messages:
    10
    Likes Received:
    0
    I will soon receive my replicape - the future brains and last missing piece of the latest addition to my printer family, which is able of monitoring the cold-end using Dallas 1-wire temp sensors. Since the hotend fan is the loudest source of sound on my printer, I was considering to actively control the fan and keeping the cold-end temperature right next to the two heatbreaks below 60°C that way.

    I read in some places that the cold-end (aka the heatsink) of e3d hotends should always be cooled with max fan power. Now I'm wondering about how dogmatic this statement is. Does anyone have experience with active cooling? Should I forget it right away, or is it worth playing around with?

    Florian
     
  2. UlrichKliegis

    UlrichKliegis Well-Known Member

    Joined:
    Sep 5, 2015
    Messages:
    187
    Likes Received:
    47
    Florian, quite frankly, I would not dare to ruin the e3d device. The temperature gradient is the essential feature of their gear. If you want to regulate the fan, you'll have to pick up the target temperature somewhere. But where? How to you know the time constants until an energy change appears at your sensor? How do you handle that for a dynamic material flow?
    In my opinion, nothing beats the cooling saturation the principle relies on. If you want to avoid the open and somewhat noisy fan, get a water cooled one like the Kraken.

    Worth playing? Always open for some fun! But make sure before you have a plan how to clean the filament duct in the cold end from molten filament.


    Cheers,

    Ulli
     
  3. triplanedave

    triplanedave Member

    Joined:
    Oct 31, 2014
    Messages:
    16
    Likes Received:
    0
    100% cooling. All the time. I fought filament feeding issues (commonly listed in the forums as "hot end jamming") with the E3D v6 for a few months. One thing that made a substantial difference is leaving the fan on 100% of the time blowing over the cold end. I just wired it up to my incoming 12V power. That way, when the possibility of heating up the extruder (e.g. because I now have 12V coming in) was always safe because my cold end was going to be cold.
     
  4. UlrichKliegis

    UlrichKliegis Well-Known Member

    Joined:
    Sep 5, 2015
    Messages:
    187
    Likes Received:
    47
    To make the 100% cooling 100%+, don't forget to keep the fan on until your hotend has cooled down to a harmless temperature - otherwise, the filament may / will melt up in its channel, and you are stuck.

    So, 100% is slightly below the real demand.
     
  5. elmoret

    elmoret Administrator

    Joined:
    Mar 1, 2014
    Messages:
    586
    Likes Received:
    91
    If noise is the issue, there are (rather expensive) 30mm fans that will do the 4CFM needed with less noise.
     
  6. florian

    florian Member

    Joined:
    Apr 27, 2014
    Messages:
    10
    Likes Received:
    0
    Okay, thanks for the info all. I will refrain from using the feature in that case and use the sensor I already bought for controlled cooling of my enclosed electronics instead.

    By the way, I used smoothie's temperature-control switch to turn on the cold-end fan when the hotend sensor reported 60° C with good results for all materials I've used (pla, abs, t-glase).
     
  7. Sanjay

    Sanjay Administrator
    Staff Member

    Joined:
    Feb 27, 2014
    Messages:
    125
    Likes Received:
    118
    I'm with Ulrich on this, it's complex and saturating cooling is the most foolproof option. Even if you got it dead right I don't think you would get much less noise as you will still need to run the fan at quite a high duty cycle to get to where you need to be.

    It's important to bear in mind that there are thermal gradients and junctions at play here. The temperature of the heatsink is not the temperature of the heatbreak at the critical transition, the heatsink runs cooler than the break for obvious reasons. From my limited playing at trying to analyse this I estimate that if the heatsink is above 40-50C then you're going to be in trouble with PLA. But as Ulrich states you're creating a dynamic system with lag and gradients if you introduce active control instead of steady state saturation. That lag might mean you destroy the critical thermal transition and jam up before your sensor notices and you can act to compensate with the fan. The heatsink has a large thermal capacity compared to the break, and I would expect that you would have lag of nearly a minute before a sensor on sink observes a rise in temperature to 60C, by which time it's far too late and you have a horrible jam on your hands.

    If you want less noise using a larger quieter fan is going to be cheaper and easier that a bunch of sensors and control systems, and it's extensively tested, known good.
     
    Mike Kelly likes this.
  8. florian

    florian Member

    Joined:
    Apr 27, 2014
    Messages:
    10
    Likes Received:
    0
    Sanjay, thanks for the thorough explanation. I'll refrain from active cooling and live with my contently humming, reliable printer :)



     

Share This Page