3
$\begingroup$

I'm trying to set up an enclosed (custom enclosure) Prusa i3 style printer for ABS, but having a fair amount of difficulty preventing the part corners from curling.

Example of curling at part corners

Detailed specs:

  • Geetech A10 (generic Prusa clone) w/ PEI heatbed
  • Custom enclosure
  • Slic3r software
  • ABS filament (obviously!)
  • 247 °C hotend temp
  • 115 °C heated bed first layer, 110 °C for others
  • Fan on for layer 3 and above

Verified heatbed temperature with calibrated IR imager.

Thus far I have been unable to prevent parts from separating from the heatbed during print, primarily at the corners where stress is concentrated. I've tried various heatbed temperatures from 90 °C to 115 °C, lower hotend temperatures (which just made the problem worse and caused complete print failure), cleaning the PEI surface with alcohol, etc. to no avail. I'm even seeing this to some extent with Benchy, it shows up as a lift to the stern and bow (slight bend parallel to the keel) -- the print is otherwise basically perfect.

I've attached an image of the more extreme curling -- yes, I should probably be using mouse ears on a part like this, but I see the same thing on parts that shouldn't require mouse ears.

What is the best way to fix this particular problem? Temperature adjustments, brims, rafts, something else?

Improve this question
$\endgroup$
6
  • 2
    $\begingroup$ I don't believe you should use a fan at all when printing ABS $\endgroup$
    – kolosy
    Commented Apr 15, 2019 at 14:42
  • $\begingroup$ I tried it both ways. Didn't see much difference, not sure which one is better in a heated enclosure. Fan in an unenclosed area would definitely be a very bad idea! $\endgroup$
    – madscientist159
    Commented Apr 15, 2019 at 14:56
  • 2
    $\begingroup$ Have you tried using a brim? It looks like you're having an adhesion issue which a brim may account for ... could also be too hot. If a brim doesn't work, I'd try lowering your hot end temp down to 230 °C and bed temps down to 105 °C. 247 °C seems quite a bit too high for ABS. $\endgroup$
    – Pᴀᴜʟsᴛᴇʀ2
    Commented Apr 15, 2019 at 16:07
  • $\begingroup$ Normally, 247 ℃ is pretty hot for ABS, the higher the print temperature, the larger the shrink effect. Use brims and an adhesive between print and bed (ABS juice, 3DLAC, Dimafix, etc.). $\endgroup$
    – 0scar ♦
    Commented Apr 15, 2019 at 20:57
  • $\begingroup$ I haven't tried a brim yet. I did already try 230 °C, the result was almost no adhesion at all with the part lifting off the bed by the third or fourth layer. The ABS in question here has a recommended extruder temp range on the spool of 245 °C - 265 °C, it's the Amazon Basics grey ABS. My extruder max temp is 250 °C so I'm a bit limited in how far I can go in the recommended range. $\endgroup$
    – madscientist159
    Commented Apr 15, 2019 at 22:47

2 Answers 2

Reset to default
2
$\begingroup$

Adding 20mm mouse ears was sufficient to resolve the problem using the original extrusion / heatbed settings. I did not expect mouse ears to be required on the Benchy model, but given the lack of better advice it seems this may be a poorly documented "feature" of ABS.

Results on the original worst-case test model:

Properly bonded mouse ear

Note that the brims will not help if the bed is insufficiently leveled -- the brim has to actually merge with / melt into the base part layer, so the extruder height has to be perfect at the brim to part interface.

Example of extruder too far away from bed:

Failed mouse ear

Improve this answer
$\endgroup$
1
  • $\begingroup$ How did you add them? $\endgroup$
    – FarO
    Commented Dec 17, 2019 at 9:30
1
$\begingroup$

A tall skirt (like 1 cm tall or more, even as tall as the part) few millimetres from the part would shield the corners and the outer parts from colder air and keep the part temperature high, reducing curling.

It is in principle better than mouse ears, because these just pull the corners, which will still have a lot of internal stress when cooled, but if you keep the part warm as I suggest, the corner will stay flat(ter) by themselves, resulting in a stronger and better print with less internal stresses.

Improve this answer
$\endgroup$

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .