20 December 2020

PROJECT: Prusa i3 MK3S Extruder Heatsink

It may come as a surprise but some 3D printer extruder stepper motors tend to run quite hot, like +25°C above ambient hot. This is not too bad when your ambient/room temp is ~20°C, but boy do the summers in Australia get toasty... and if things get toasty enough the heat can creep through the shaft/driver gears and warm up the filament surface, making it that much harder to feed in controllably

Now as far as I am aware I have not run into this issue with my Prusa i3 MK3S + MMU2S, however I was curious to see:

  1. What is the typical temperature rise of the extruder stepper motor
  2. If installing a passive (no fans) heatsink could help lower the above temperature

The Setup

I tested the extruder stepper motor of my Prusa i3 MK3S under 3 scenarios:
  1. Original (no heatsink)
  2. Dense fin pattern heatsink (ATS-FPX040040025-05-C1-R0)
    • Single fin dimensions are 0.8 x 8.5 x 22.2mm and there are 50 of them
    • Weight 29.6g
    • Thermal resistance 8.7°C/W (unducted flow)
  3. Light fin pattern heatsink (ATS-CPX040040025-116-C1-R0)
    • Single fin dimensions are 0.9 x 8.5 x 22.2mm and there are 32 of them
    • Weight 21.4g
    • Thermal resistance 4.1°C/W (unducted flow)
    • Suspect this will be the winner due to the spread fins 

In each case I logged the stepper motor & room temperature, from which the temperature rise (above ambient) value was calculated. I should point out that comparing the Δ/change in temperature is more suitable here instead of say comparing the absolute peak temperature in each scenario 

Also to make sure I was testing the heatsink performance (and not how well it coupled to the stepper motor) I used a crazy high thermal conductivity pad. The EYG-A091202DM, which is graphite based and has a thermal conductivity of 1850W/m·K! (two orders of magnitude higher than your typical thermal pad)
NOTE: To mount the heatsinks I had to change the M3 screws from 30mm to 35mm

Finally, for those curious this is the model I was printing for the test:

The Results

As I suspected the lighter fin pattern heatsink (ATS-CPX040040025-116-C1-R0) gave the lowest temperature rise, reducing it by a cool 7°C compared to no heatsink

On paper this may sound surprising, as this heatsink has a lower thermal resistance (4.1°C/W vs 8.7°C/W) which in theory should mean worse performance. BUT what one must realise is that we are extracting/exchanging the heat from the heatsink via convection currents (not forced air flow), and in this case it's easier for the air to couple into a less dense fin pattern


  1. wow amazing it actually lower the heat

    1. Hey there. Sorry for the late reply, my spam filter was a bit too strong...

      Yea, was also interesting to see that more fins does not mean better cooling

    2. have you thought of trying fins oriented horizontally instead of vertically? since the head is moving along that axis, head movement should be a bit more effective in leaving out the heat that way, though with smaller prints it might not be

    3. Yea like you said, won't make much of a difference unless you are moving horizontally at great speeds. So suspect you get way better results from current orientation (think convection currents from build plate)