Layer delamination with ASA

[LindsayPatten 1]

I thought I had figured out how to avoid delamination when I was printing half-size prints of this model in ASA, but when I printed it full size (246mm height) delamination reared its ugly head again:

I increased the bed temperature to 105 and decreased print speed from 125, 100 to 100,75 (inner wall, outer wall) which decreased the number of delaminations slightly but...

I'm printing at 0.2 layer height, 270 degrees and keeping the ambient temperature at a steady 41 degrees.

It's an airfoil shape with a fairly sharp trailing edge which is the hardest to avoid delamination with but it also starts on the leading edge.

The filament label says 240-260 but I couldn't avoid delamination on the smaller prints with less than the 270.  I might try 275 and/or 0.15 but I'm wondering if there is anything to do with infill that might help.  Or adding some additional reinforcements or something.

Any advice?

 

[Slashee_the_Cow 421]

I think Grid infill is supposed to be the strongest in the vertical direction, so you could try that. Also try increasing Infill > Infill Overlap to make sure the infill is sticking into the walls. Walls > Alternate Extra Wall also helps the walls grab onto the filament (since it'll sandwich it between lines) but obviously adds weight and filament use.

 

Also not that I've had problems with delamination, but I think I read somewhere that if you turn off (or at least turn down) the cooling that can help prevent it.

 

And don't forget Slashee's Silver Rule™: Small scale testing is your friend.

[LindsayPatten 1]

As it happens I am using Grid infill already, although it is interesting that when I used Lightning infill there wasn't any delamination.  I'm still chewing on that.

I've actually got the model and side fans turned off and have the back (vent) fan at 10% to keep the ambient temperature around 41 deg.

Wrt small scale testing that's exactly what I did, although it still took me almost two reels of filament to figure out that even though the filament guide says 240-260 that it actually needs to be 270, and that even though I've never used glue for printing PLA or PETG, for ASA prints of any size glue is essential.  I no longer have problems with small scale stuff but when I print a piece that takes most of the diagonal size of the printer I'm getting serious delamination on the front and back tip.

I realized this morning that I had the z-margin on the trailing edge, which is aesthetically the best but possibly contributing to delamination on that edge.  Before I had that insight I created and started a print with an extension of the trailing edge out to a cylinder which should be less delamination-prone and is easy enough to remove post-printing.  I printed off a partial height section of the blade with that and it seemed to work.  In another 10-20 hours we'll know if it works for the full height piece.  I printed the test at 275 deg on a 105 deg plate with reduced speed, and reduced the layer height to 0.16 from 0.2, so we'll see if that combination works for the full piece.

[LindsayPatten 1]

As a follow-up in case anyone else runs into a similar scenario:

 

After reading more about the causes of delamination/layer separation I found out that residual stress (aka internal stress) caused by material shrinking as it cools scales with the size of the print.  For any given set of number of perimeters, infill, etc., there will be a maximum size that you can scale the model to before residual stress will cause problems.  As a model gets taller the heated build plate can't heat the whole height of the print so that doesn't help.  So it's not surprising that even when I got everything working at half scale there were problems when I printed at full scale.

 

In my case increasing the wall line count was how I got a successful print.  Increasing wall count for the whole model added too much weight so I used support blockers with per model settings to only increase it at the leading and trailing edges.  The model has 3 wall perimeters generally and I still got a small delamination  adding 3 additional wall layers on the leading and trailing edges so I gave 5 additional walls a try and that was successful.  I should try 4 additional walls but each model takes almost half of a 1kg reel and I've wasted multiple reels already so I'm sticking with 5 for now.  I'm still a little surprised that three additional walls, for a total of six, was insufficient but I guess it goes to show how strong residual stresses on large parts can be.  The model is 247mm tall and 358mm long.

 

I also added two-layer "mouse ears" at the bottom to combat warping away from the build plate as brims alone weren't strong enough to stop that.

 

Edited September 8 by LindsayPatten
Typo