Blobs on layer change

On the place for external perimiter start and stop there forms a small blob.

What should I start to look at in slic3r?

Mendel90
E3Dv5
Slic3r 1.2.9

Cheers
Tobbe

Increase your retraction length and/or speed, and also try lowering the temperature.

Tobbe,
You can also try turning off Retract on Layer Change - in Print settings - Extruder 1.

The blob at the end is caused when the retraction happens while the nozzle is stationary.
This blob is worse with Bowden extruders because the retraction length is longer, and so has more time for the ooze blob to develop. So extra retraction could also cause it.

This was discussed in this thread

---

My printer: Raptosaur - Large Format Delta - [www.paulwanamaker.wordpress.com]
Can you answer questions about Calibration, Printing issues, Mechanics? Write it up and improve the Wiki!

Thank you for the input.

I have now tried more retraction, no retration on layer change and lower temp. I have tried them seperatly and together. The blobs are still there.
This is a new problem that I am not really shure when it started. I have not done any changes in the slic3r settings.

So if it is not a setting, what else should I look at? any idéas?

How much retraction are you using? Do you have a Bowden extruder, and if so, how long is the Bowden tube?

---

Large delta printer [miscsolutions.wordpress.com], E3D tool changer, Robotdigg SCARA printer, Crane Quad and Ormerod

Disclosure: I design Duet electronics and work on RepRapFirmware, [duet3d.com].

I solved the problem.
Sliced the model with version 1.1.7 of Slic3r instead of 1.2.9! :-P
Seems there is a bug in the newest version that have a problem with layer change on thin walls.

It would be helpful if you could post the gcode for both to help determine what is happening.

This file is sliced with Slic3r 1.1.7
117.G

And this is the blobby one from 1.2.9
129.G

I found I had the same problems and couldn't get rid of them no matter what I tried. I had been using Slic3r and decided to try another program. First time I sliced the same file with Cura they disappeared.

Tobbe,
Thanks for posting the two files. I am comparing the two and see that 1.2.9 is extruding a lot more.

I'm sorry I wasn't more specific - my debugger relies on Verbose Gcode to peek inside Slic3r. Can you turn that on and re-slice in each version?

To turn on verbose gcode: Print Settings, Output Options, Verbose G-Code checkmark.

Thanks
-Paul

Ok, another go this time with verbose G-code

1.1.7: 117.G

1.2.9: 129.G

The STL: test.STL

The two outputs are quite different, the 117 file has 2 perimeters, and the 129 has 2 perimeters and an infill.

There are some short infill segments that are coded to print at over 107 cubic mm3/second. Not sure yet how that may be involved. I will be analyzing these more tonight.

I've gone through both gcode files with my debugger, line by line.
As expected, these two gcode files are very different, although the Slic3r settings appear to be the same, and both have Seam Position: Random.

Version 1.1.7 is only doing two perimeters, even though "Perimeters = 3", and I note: it does not always even do two perimeters for a few sides. The relevant width settings are:

  ; perimeters extrusion width = 0.40mm
  ; infill extrusion width = 0.42mm

- The start and stop point for each perimeter is always right near the corners. That is one reason you do not see any blobbing in the center.


Version 1.2.9 is doing one internal perimeter, one external, then one infill. The extrusion width for infill is much larger than in 1.1.7. The relevant width settings are:

  ; external perimeters extrusion width = 0.40mm
  ; infill extrusion width = 0.48mm

I've numbered this block of gcode from layer 3, and added some detailed comments to each line.

 1  G1 X-17.300 Y-11.442 F7800.000          ; 14.553mm move to first perimeter point
 2  G1 E1.00000 F1800.00000                 ; unretract
 3  G1 X-17.300 Y-17.300 E1.09245 F831.497  ;  5.858mm perimeter  16.487 mm3/s
 4  G1 X17.300  Y-17.300 E1.63854           ; 34.600mm perimeter   1.395 mm3/s
 5  G1 X17.300  Y17.300  E2.18463           ; 34.600mm perimeter   1.395 mm3/s
 6  G1 X-17.300 Y17.300  E2.73072 F831.497  ; 34.600mm perimeter   1.395 mm3/s
 7  G1 X-17.300 Y-11.382 E3.18342           ; 28.682mm perimeter   1.395 mm3/s
 8  G1 X-16.954 Y-11.642 F7800.000          ;  0.433mm move inwards before travel
 9  G1 X-17.050 Y-11.382 F7800.000          ;  0.277mm move to first infill point 
10  G1 X-17.039 Y-17.039 E3.29485 F791.902  ;  5.657mm infill      1.658 mm3/s

- Line 3 is the start of the first outside perimeter. Note that the extrusion rate is very high: 16.487 mm3/s for that short segment

- Line 7 is the last outside perimeter.

- Line 8 and 9 are two short moves - and there is no retract there before starting infill.


At other times 1.2.9 does not have the unretract at the first outside perimeter point (which is odd), and there is no over extrusion called for like line 3 above. It still has the two short moves like line 8 and 9 above. This is the outside perimeter on layer 8:

G1 X-7.418  Y17.300  F7800.000           ; 15.701mm move to first perimeter point
G1 X-17.300 Y17.300  E3.27626 F831.497   ;  9.882mm perimeter  1.395 mm3/s
G1 X-17.300 Y-17.300 E3.82235            ; 34.600mm perimeter  1.395 mm3/s
G1 X17.300  Y-17.300 E4.36844            ; 34.600mm perimeter  1.395 mm3/s
G1 X17.300  Y17.300  E4.91453            ; 34.600mm perimeter  1.395 mm3/s
G1 X-7.358  Y17.300  E5.30370 F831.497   ; 24.658mm perimeter  1.395 mm3/s
G1 X-7.618  Y16.954  F7800.000           ;  0.433mm move inwards before travel
G1 X-7.358  Y17.050  F7800.000           ;  0.277mm move to first infill point
G1 X-17.039 Y17.039  E5.49442 F791.902   ; 9.681mm infill  1.659 mm3/s

In my opinion the problem may be threefold:
1) Starting the extrude in the middle of a line
2) Overextrude in the first segment
3) The two moves right by the start/end point could allow a bit of ooze.

Where to go from here:
-If you can attach the STL file, I will post an issue on Github, or feel free to do so yourself. You could link here.
-You can try a different Seam Position setting.

I've attached the commented Gcode my debugger produced.

Edited 1 time(s). Last edit at 09/05/2015 10:27PM by Paul Wanamaker.

---

My printer: Raptosaur - Large Format Delta - [www.paulwanamaker.wordpress.com]
Can you answer questions about Calibration, Printing issues, Mechanics? Write it up and improve the Wiki!

The STL is attached in the message with the verbose G-code files :-)

My solution to the problem is to use 1.1.7 as there is nothing in 1.2.9 that I miss.

Impressive analysis Paul.

Thanks James.

Tobbe can you please post your config file from Slic3r 1.2.9, and which operating system. These are items I will need in order to post the Issue.

Thanks

Windows 8.1
Config.ini attached

I've fought with this issue before, I simply solved it by performing retraction along a seam, slic3r has that option somewhere.

I have seen similar blops with ver 1.2.9 when travelling but not found combinations that makes it retract during travel.
Going back to ver 1.27 seems to be different and ok, you may try it.

I have to correct this one as I have verified that even 1.2.9 retracts correctly, but somehow problem is with fill in extrusion.
When travelling from fill in to perimeter pressure seems to be very high in extruder so even retraction is fired it still makes a bobble at the perimeter where landed.
it would need some fifling with fill in volume to find out more abou this.

Edited 1 time(s). Last edit at 10/24/2015 05:51AM by justcurious.

Quote

performing retraction along a seam, slic3r has that option somewhere

Would like to find where this option is.