Seam/Z-scarring significantly better in Silc3r - Settings?

[Malakian]

Hi,

So I'm having some bad seam/z-scarring on S3D, have been playing with all kind of settings for two days now without any note worthy improvement.
Then for the hell of it, I thought I just try Slic3r instead. And to my joy/dismay it's tons better with slic3r, even with the exact same settings. Or as best as I could match them. For some reason though Slic3r only gives me 1 perimeter even though I'm asking for two, not sure what's up there.

I was wondering if anyone could possibly look over my settings files and see if they are truly the same or I'm missing something?
I really hope it's a setting issue, because I love S3D in all aspects but those seams are horrendous. And kinda sucks spending a lot of money and getting better results with a free software, as of now anyways.

Setting Files: https://drive.google.com/open?id=0B5qau ... 0NjMV96XzQ

Edit: Looked over the setting files, they don't match completely. Did a little tweaking in slic3r since the seam would just be pin holes on default settings. And had to up the mutliplier by 0.01 in Slic3r to get solid layers. Kinda weird. Default Slic3r, pin holes no blob. Default S3D, divot and blob.

The blob is at start of the layer. Tried No retract, 0.8mm at 80mm/s, 0.8 at 20mm/s with default jerk and acceleration (its a rigidbot, default is really low). Tried same retract settings with over doubled acceleration and jerk. Try with wipe, from 0.1mm-5mm, Coast from 0.05-0.20. Negative restart distance (that just lead to holes and bad first layer). With forced retract on layer change, and without. And countless combinations of the values mentioned. Nothing seem to do much at all.

Best print so far with S3D is pretty much default ooze/retract settings. Just turned off Only retract over open spaces.

Print settings:
PLA @ 190c/60c
speed: 60mm/s

It's really hard to get clear pictures especially white, but you can see the difference. Also the divot is deeper and the blob bigger on S3D print. By quite a lot and not as neat. I'll try to print in another color so it shows better in pictures.
Pictures: http://imgur.com/a/9b0cL

Thanks in advance for any tips!

[Malakian]

Update: I'm playing around with Slic3r to try and get the best results I can, then transfere those settings to S3D. Since it seem to work better there.

Also, while playing with the Extra restart after retract settings, I manage to recreate the seam I get in S3D, or at least very close. This was with a 0.30mm (high) positive extra restart distance, so logically that would mean I should use a negative restart in S3D to get the same results as Slic3r without Extra restart. But no. Curiouser and curiouser

Once I get it as good as I can, ill post some new pictures with a better color.

[wirlybird]

Thanks for the effort.

I have the same issues in S3D. At the start of a line it spurts out a blob of material. I have tried a negative restart with some success but it has consequences in other areas.

[Malakian]

Yupp, that's exactly what happens with my prints too. If I start playing with Extra Restart distance it affect other areas too.

Hopefully it just a settings issue, will update as I discover more with trial and error.

Seems I got the best I can do atm on Slic3r, changed to PETG which is more oozy than PLA, hopefully making the seam more dramatic showing the issues better. Also it's opaque bright green so it shows in the pictures.

Just doing some comparison prints on Slic3r and S3D right now. A seam torture test and a benchy.

Edit: That's really weird. So I made sure the settings matched identically this time. And instead of printing through S3D I just sliced it and saved the Gcode. Printed it trough Pronterface, since that's what I use for Slic3r. So that is not interfering with anything. Weird part is, Slic3r printed the Seam torture in 16 min, S3D estimated at 37 min. Not done yet though so not sure if its actually slower or just calc. error.

If it is indeed slower that would explain the poorer seams. As you want as little dwell time as possible on retract/layer change. So slower means more pronounced seam, at least in this case. But I can't find any discrepancy in the speed settings on the two slicers

Edit2: Jupp, It actually took longer than Slic3r, so I stopped it after 16min. And the seam is REALLY bad compared. Benchy's next, weirdly enough both those come out estimated Slic3r: 1hr 01min, S3D 1hr 03 min.

Settings file used this time for both prints (_2): https://drive.google.com/open?id=0B5qau ... 0NjMV96XzQ

[Malakian]

So, here are the results.

PETG
230c/70c
60mm/s
0.20mm layer
20% infill.
Benchy actual print time on S3D: 1hr 08min. Slic3er: 1hr 06min.

Overall S3D did a better job. Slic3r had some heat issues, but the seams are night and day (Slic3r is way better). Can't really find the seams on the Slic3r model except on the doorway, while the S3D model is quite easy to find the seams, and the hull have sits/divots here and there which Slic3r does not. I'm at my wits end, I don't know what more to try. Seems like I just have to live with the seams if I want to use S3D.

Doing a last Benchy on the Slic3r Prusa Edition. People are saying good things about it, and from what I've read it almost seem like a cross between Slic3r and S3D. We'll see if that does any better. If I could just have the interface and overall quality of S3D and the Slic3r's seam I would be a happy camper.

Seam Torture : http://imgur.com/a/WqOzO
S3D top
Slic3r bottom

Benchy -
S3D: http://imgur.com/a/5YI7l
Slic3er: http://imgur.com/a/uMJIF

Edit: Slic3r Prusa edition did pretty well, but seem to be still some bugs. Some weird artifacts some places and a lot more stringing. But still better seams than S3D

[wirlybird]

60mm/sec seems pretty fast for PETG.

The Prusa version of Slic3r is optimized/rewritten for the MK2 printer.

Interesting results though.

[Malakian]

Might be a tad to fast for optimal results. Much slower though, 40mm/s and lower. And the surface finish gets fuzzy with PETG for some reason. But I had the same with PLA. Just wanted to use PETG, since it oozes more it will show the seam defect more.

[Malakian]

Hmm, I just noticed. S3D has a own setting for Z Axis movement speed. While Slic3r just has Travle speed (which I guess uses the same for XY and Z).

Might be something there, have to try that. Seeing as this issue is Z related. Little hesitant though, default is 16.7mm/sec, and XY is 80mm/s. Quite a substantial speed increase to up it to 80mm/s.
Will have to wait until tomorrow though, printing a Yoda at .10 layer height. Still some hours to go.

edit: Found this while searching Z axis movement speed. " believe the max speed defined in the firmware for the Z-axis is 250 mm/s (or 15000 mm/min) and the max acceleration is really low at just 30 mm/s^2 (vs 9000 mm/s^2 for the X/Y axes!). Even if you tell the Z-axis to move at a fast speed, the firmware will step in and limit it to those values. "

So I could set it 1000mm/s and the printer shouldn't wreck it self, as the firmware would not allow it. If I understand it correctly. Guess its only one way to find out. Wish me luck

Also just did a search on "Slic3r Z travel speed". Suddenly quite a lot of hits on people struggling with too fast Z movement when using Slic3r. Seems like I might be on to something. Yoda might need to be sacrificed, to excited to try it out.

[Malakian]

Sadly, Yoda is no more, or he's missing the better half of his head.

BUT, eureka! Seam torture is now pretty much identical to Slic3r's
And this also explains the slower printing speed with S3D on the Seam torture, since it's a lot of z movement on that model. It all makes sense now.

Man, so typical my luck. Literally the only setting I have not fiddled with, and that's the culprit. Meh, I'm just happy I got what I wanted. Can still use S3D and got my "Slic3er" seam's

Hopefully this might be of help to you too Wirlybird. If not, if you give me more details I could try and help, starting to get pretty good at this seam troubleshooting thing

Proof: http://imgur.com/DnneHmS Pretty much identical, might even be a tad better with S3D now. Might be biased though

[wirlybird]

What setting did you change? I'll try to find a pic of issues I have had on a cylinder and post later.