Edges in prints are curved

[JeffB]

I'm new to S3D, and am trying to dial in my S3D settings. I've created a simple 20mm x 20mm x 10mm box in 123D Design to try an calibrate and test my system. I've leveled my print bed, but I'm seeing an issue with the edges/sides curving... when the print is finished, the vertical edges are curved. The model is a box, with 90 degree corners, so I wasn't expecting the results that I'm getting.

Here's my setup:
Windows 7 Pro communicating over USB to my printer
Taz4.1 printer (stock, w/no firmware or other hardware modifications)
Printing ABS I purchased from Lulzbot

Before I print, I pre-heat up the bed to 91, and the nozzle to 231, and when I print I have the "wait for temperature control to stabilize before continue" option turned off.
(and fan speed is 0)

For the S3D Process settings, I'm:
- using the default Taz4 process created by S3D
- selected the "auto configure for ABS" and the "auto configure for High" settings
- changed the layer's top and bottom solid layers to 10 each
- changed the number of outline/perimeter shells to 4
- all options in the Additions tab page (skirt, raft, etc...) are off

I thought the issue might be due to be height and bed leveling, so I ran through bed leveling. I also played around with the bed height, but I think I have it dialed in.

Any idea as to what's going on, and how to fix it?

I've played around with the S3D settings, but haven't been able to figure out what the issue is.

The only thing else which comes to mind is that I'm not printing inside an enclosure. There aren't any fans blowing on my printer (the fan on the power supply is directed away from the bed/nozzle). I have everything set up in my attic, and the ambient temperature right now is around 64 F, though I didn't have this kind of problem printing edges with Slic3r and PrintRun.

Thanks for any help,

Jeff

[MikeO]

While folks say you don't normally run the cooling fan when printing ABS, I find I have to do that, at about 50% or 60% fan speed, to keep that effect from happening.

Also, I usually heat the bed to between 100 and 110, for ABS.

[JeffB]

re: temp: I tried printing 240/110, as well as 230/85, and both exhibited the same issue.

re: fan: Prior to S3D I was using Slic3r and PrintRun, and ran into all kinds of problems when the fan was on. Problems like delamination, parts curling off the bed, etc... I tried turning off the fan (when printing ABS) and the difference was night and day, so I don't ever use the fan.

[MikeO]

I understand about the problems with running the fan with ABS, I've had those myself.

But running the fan at 50% or 60% (which feels to be WAY less airflow than 1/2 of 100%) seems to avoid those problems, while helping the curving. At least for me, anyhow. I'm printing on a Lulzbot TAZ5.

The cooling fan and ducting on your machine may be very different than the one on mine, though, as far as what "50%" really means in terms of airflow.

EDIT: To help avoiding curling, I set the fan speed to zero for the first two layers. Then I set it to go to 60%.

[chriswh86]

I am running into the same issues. Usually with PLA as that's all I have been printing lately. I've been doing some testing tonight as I want to get it worked out before I print some models.

I played with the z-axis offset a bit, i went to .02, .05 and finally .08. I don't believe this actually helped. I then kept the z-axis offset at .08 and then changed the first layer width from 100% to 75%. This made a large change. Nearly straight up on the side but not quite. Now I am at 50% first layer width and I have the z-axis offset back at 0. I made the large change to 50% just to see that its really what is doing it.

I have run into issues in that at 50% first layer width and no z-axis offset and a concentric external fill pattern, after about 3-4 rings in the extruder starts depositing PLA partially on top of layers that have already been down, ex. layer 5 is partially covering layer 4 and layer 6 is partially covering layer 5 and so on. It straightened its self out about half way in but I still don't like it. I am assuming this is due to the first layer width being 50%. (ive never used z axis offset until tonight in the above tests). I just pulled the 50% first layer width sample off and its pretty ugly on the bed plate side. It bonded fine for the print but there's definitely a rough bottom. Not your typical no gap glossy bottom.

I am now trying 70% first layer width and still keeping the z-axis offset to 0. It appeared to run over its previous layers a bit but not as bad as the 50% version. I will see how it turns out. Unfortunately I will still have a less than optimal build plate side of the test, as my 75% test showed a rough bottom as well.

I am not sure if now I should be trying some z-axis offset and some first layer width adjustments? Id like to have a good bottom layer, good adhesion but I am not sure what steps to take next to produce a model that's basically straight up on the sides...

Hopefully a bit of this helped JeffB and hopefully some others can assist us further...

Lulzbot

[JeffB]

Hopefully a bit of this helped JeffB and hopefully some others can assist us further...

Lulzbot

Hi Chris, sorry for the late reply - I never got notice of your reply (I thought I had reply notifications turned on, but I guess not).

So I kind of have the issue worked out, but I'm still tweaking it. The issue (at least for me) had to do with print speed. I'm not in front on S3D right now, but S3D has a setting... something along the lines of "if time to print layer is less than X seconds (I think 15 seconds was the default), reduce speed to 20%"

To force the speed down, I set it to something I knew would trigger it, so I change it from 15 seconds to 150 seconds. My test print (a simple 20mm x 20mm x 10mm block) printed DOG SLOW, but it worked - no curling.

I then changed the profile back to 15 seconds, and then started to play around with reducing the overall nozel speed. I'm still trying to dial it in, so I'm still performing A/B tests, but hopefully this will point you on the right path.

Jeff

[chriswh86]

I will give that a try and look forward to seeing what else you find.

I did try a .3mm chamfer on my most recent part with perfect results but its a crutch to the actual issue so I will keep looking