Two corners bad, two corners ok

[konghoei]

Help! New to 3D printing and still trying to learn. I have this problem on an enclosure that I am working on. Two corners were ok, two corners have problem shown in the attached image. I can't figure out why. I tried Wipe as well as Coast for 0.5mm, didn't make a difference.

Simplify3D. Some params that may be of interest below:
0.2mm Z-res
Hairspray on heated bed @ 70C
Fan @ 100% from layer 2.
1mm retraction distance, 2400 mm/min rertaction speed.
Force retraction between layers: selected
Only retract when crossing open spaces: not selected

Any suggestion on what I should try to fix this?

Thanks
H.

[Tim]

Difficult to see from the picture---are the corners lifting up? Is there an overhang on the side? What type of filament are you printing in?

Lifting at corners is pretty common for large, flat objects with walls, like enclosures. Enclosure bases are like worst-case scenarios for shrinkage. PLA ought to be okay, but ABS and PET will shrink until something is forced to give way. You can add a wide skirt, or at least "mouse ears" at the corners to help keep the corners down, but shrinkage creates a LOT of force. . . there are a number of people who have found that a large enough ABS print with a strong enough adhesion to the glass will simply break the glass. I'm sure there must be design techniques for alleviating the stress from shrinkage, but I'm not a mechanical/civil engineer. I tried a few things on my Smoothieboard enclosure design, but I don't think they were particularly effective.

[ErikAkia]

It is hard to see but is the edge on the left at an angle for the first 20 or so layers, like a chamfer?
If it is I would say this looks like curl. Where the edge is being printed further out from the one below and it curls up. Then when the next layer is put down further out it will either push down the curl but it deforms that layer and the curl bounces back up or it goes around the outside of the lifted edge and droops and sags with no support.

First thing to try would be to make sure you are printing the perimeter inside to out. This can affect surface quality a small amount but when you print over hangs it can make all the difference.
Plastic shrinkage plays a big part and if this is ABS try switching to PET(G or +)
Slowing down the print speed should help. You could try slowing just the perimeter speed more. I have not tried that yet.
I think over extrusion plays apart so you might dial back your extrusion multiplier a small amount.

[jsc]

It doesn't look like pull up. My guess is you are printing in PLA and the two bad corners are the ones in the rear. The other two corners are okay because the bed fan is blowing directly on them. Try putting a desk fan at the back of the printer blowing forward. You can also try printing a shroud for your extruder that will direct more airflow to the print area. This only applies to PLA.

[PcS]

For a .2 layer height those layer lines look fat to me. I wonder if z height and bad bed level caused this....are the two corners opposite diagonally from each other that are doing this ?

[konghoei]

Thanks for all the replies and feedback.

The material is PLA. Overhang is at slightly above 45 degrees.
I have tried with several different orientations and, yes, the corners that curled the worst are usually the ones that are in the rear.

I will try slowing down the perimeters and adding a small fan near the back. Will also try adding brim or just put one of them corner ears.

Thanks!

Side note:
Yes, I have been having problem with the HPB being about 0.2mm higher near dead-center. When I clip the glass bed to the HPB then the glass seemed to curve down from the middle out. This makes bed leveling and printing the enclosure much harder. I had to print it at one of the corners and try to avoid the center area. And to compensate for the difference, I had to increase the extrusion rate to find a balance between one side having too much and the other side having too little. Picture attached.

I have several prints where the corners didn't lift/curl for the first few layers (with thicker and timely application of hairspray), but the higher layers still curled.

[insta]

Silly question, how many clips are you using and what order do you put them on in?

I run 3, 2 in the rear corners and 1 in the middle. The rears aren't tucked up flush against the glass, but rather against the edges of the spider with about 3mm between the glass and inside edge of the binder clip. This is enough to let the glass creep once it heats up.

[konghoei]

I used 4 clips, 2 on left (coord ~ Y=0, X=10 and 190) and 2 on right side (coord ~Y=250, X=10 and 190).
The clips inside edge flush against edge of glass.

I usually do two clips, diagonal pair, at about the same time. Then I do the next diagonal pair.

[Bluesoulx12]

I used to have the same problem as you. I've had a ton of success with elmers glue double layering. I am a mechanical engineer and i always thought it was weird that the adhesion worked for one side and not the other. I even leveled the bed a couple times just to experiment. I let the first elmers glue layer go down and completely dry (for like an hour) and then the second layer I slather that on. Keep in mind that everyone else is right on the glass chipping if you adhere a lot and have a large print so that is a disclaimer. It's risky, but it works out for me atleast.

[konghoei]

Tried the glue but I just can't get a nice even layer and that flat first layer just didn't look good at all. So I ended up with hairspray that gives me a nice smooth first layer.

So I re-orient the enclosure 45 degree (which I think changes the corner that get the least airflow to the one on the front now), add 1 layer of brim, and slowed down outline speed by to 50mm/s (was printing outline at 85mm/s). My printing time almost doubled but the result looks much better thought still have the same problems, just not as bad.

Now I need to figure out a good way to add a small fan and play with extrusion rate.