Retracts too long with 3.0.2?

[pbmax]

Has anyone else experienced retracts that are too long with 3.0.2?

I keep a bit of foam on the filament as it feeds into the extruder motor on my printer which makes it easy to tell how far it retracts. On a recent 3.0.2-generated print I watched it retract double the 7mm retract and then only feed 7mm back in (or so it seemed to my eyes - double makes more sense than some arbitrary number). After feeding half back in it would extrude normally for a bit and then sometimes it would randomly feed that missing 7mm back suddenly, nicely fouling my hobbed gear. I started watching it after losing 4 S3D-generated prints today, all about the same point with the same object, even with different scaling. The middle of the object gets really thin due to the retract issue. This object printed fine with 3.0.1. I haven't located the buggy spot in the gcode yet; I hope to find some time tonight to dig through it.

I'm using wipe while retract with a really big value for "maximum allowed detour factor" - 100. I set this a while back because I wasn't seeing it avoid outlines like it should have been and I just left it that way.

Has anyone else seen this?

[pbmax]

Here's the offending g-code; note the lines in bold. The first pair of bold lines drives the filament back to 0 after an earlier retract and resets the E axis reference point to zero. The second pair of bold lines appear to be a retract while wiping... the first line is consistent with other retract-while-wiping lines earlier in the file, but with abbreviated length - 4.8 vs the previous retract-while-wipe lines which were all E-7. And then there's the nasty line that retracts 11.8 right after that.

I didn't find any instances of normal extrusion between huge retract and replace to 0 so that may have just been my imagination. I just verified that g-code produced w/o the "Perform retraction during wipe" enabled does not have this issue. I will be disabling this!

Edit: This problem does not happen with 3.0.1. Maximum retract distance in the g-code with retract set to 7mm is E-7 with 3.0.1. 3.0.2 is all over the map; I just found a retract of E-17.5796 with retract length still set to 7!

Edit: A factory file replicating this problem is attached.

; outer perimeter
G1 X-16.842 Y-6.956 F12000
G1 Z2.200 F12000
G1 E0.0000 F2700
G92 E0
G1 X-16.842 Y-8.956 E0.0998 F600
G1 X-14.842 Y-8.956 E0.1996
G1 X-14.842 Y-11.556 E0.3293
G1 X-16.842 Y-11.556 E0.4291
G1 X-16.842 Y-13.556 E0.5288
G1 X-19.442 Y-13.556 E0.6586
G1 X-19.442 Y-11.556 E0.7583
G1 X-21.442 Y-11.556 E0.8581
G1 X-21.442 Y-8.956 E0.9878
G1 X-19.442 Y-8.956 E1.0876
G1 X-19.442 Y-6.956 E1.1874
G1 X-16.942 Y-6.956 E1.3121
G1 X-16.842 Y-6.956 F600
G92 E0
G1 X-16.842 Y-8.956 E-4.5000 F1200
G1 X-15.731 Y-8.956 E-11.8000
G1 X-14.842 Y-8.956 F1200
G1 X-14.842 Y-9.956 F1200
G1 Z2.500 F12000
; support
G1 X-15.082 Y-4.256 F12000
G1 Z2.200 F12000
G1 E0.0000 F2700
G92 E0

[JoeJ]

I have a guess at what's happening, but I wasn't able to find the bold line you mentioned in the gcode file that was created using the factory file you posted.

Can you use the factory file you posted, create gcode from it, and then post the gcode and let us know what line you are seeing the error on? I wasn't able to find it just by looking through your files

[pbmax]

Hmm, I did verify that the issue was reproduced with that factory file before I posted it, though the exact line may be different. Essentially any retract beyond E-7 is the issue; I caught it via:

grep "E-" blah.gcode

I'll upload a .gcode with the issue when I get home tonight as I don't have access to it presently.

[pbmax]

The gcode is attached.

Lines 1286 and 1287 have the first wipe-retract with the issue in this file:

G1 X97.300 Y95.700 E-4.5714 F2100
G1 X95.175 Y95.700 E-14.8367

[checktest]

I have also seen this issue. I've been getting jams on anything that required a decent amount of retracts. I noticed that sometimes I would get a standard retraction distance, and then sometimes I'd get what looked like retraction distances of double (or more?) my normal retract.

Unchecking "perform retraction during wipe movement" made the extra long retractions stop (which helps avoid jamming, but makes my print quality drop).

Is this a bug? Shouldn't the retract distance during a wipe be the same as the wipe distance? Or does it essentially do a double retraction?

[pbmax]

Yeah, this is a bug. I contacted support directly on it; they recommended checking relative retraction in the gcode config if your printer supports it. Mine do and this made wipe-during-retract functional again for me.

[checktest]

That seems to have helped, thank you for your reply!

[jamezracer]

I can confirm the same thing. In particular I notice it happens when the wipe must go over a curved path. Printing cubes doesn't give issues.

an odd retraction for me looks like this:

Code: Select all

G1 X146.785 Y35.100 E0.7375
G1 X146.697 Y34.575 E0.7482
G1 X146.656 Y34.044 E0.7588
G92 E0
G1 X146.662 Y33.508 E-0.3348 F2400
G1 X146.682 Y33.249 E-1.8363
G1 X146.759 Y32.719 E-2.8212
G1 X146.880 Y32.200 E-4.4934
G1 X147.048 Y31.691 E-6.1594
G1 X147.146 Y31.450 E-7.6612
G1 X147.379 Y30.968 E-8.6458
G1 X147.652 Y30.511 E-10.3178
G1 X147.964 Y30.079 E-11.9820
G1 X148.315 Y29.674 E-13.6481
G1 X148.317 Y29.672 E-14.9901
; inner perimeter

With relative extrusion distances enabled I get this for the same piece which makes more sense:

Code: Select all

G1 X146.785 Y35.100 E0.0106
G1 X146.697 Y34.575 E0.0106
G1 X146.656 Y34.044 E0.0106
G1 X146.662 Y33.508 E-0.3348 F2400
G1 X146.682 Y33.249 E-0.1625
G1 X146.759 Y32.719 E-0.3348
G1 X146.880 Y32.200 E-0.3328
G1 X147.048 Y31.691 E-0.3348
G1 X147.146 Y31.450 E-0.1624
G1 X147.379 Y30.968 E-0.3348
G1 X147.652 Y30.511 E-0.3328
G1 X147.964 Y30.079 E-0.3328
G1 X148.315 Y29.674 E-0.3349
G1 X148.317 Y29.672 E-0.0025
; inner perimeter

Note that when you check the box for relative extrusion Simplify puts the following BEFORE you startup script

Code: Select all

G90
M83

M83 means your startup script will now be executed with relative coordinates but you probably wrote it in absolute coordinates in the first place. I ended up filling my startup script with G92 E0 for every extrusion so it would execute the same in absolute or relative mode.

My new startups script for ultimaker 2:

Code: Select all

; note G90 and M83 are actually called before this script
; absolute XY coordinates and relative extrusion
M907 E1500; set extruder stepper current
G28 ; home all axes
G1 X0 Y5 F9000 ; bring extruder to the front
G92 E0 ; zero the extruded length
G1 Z12 ; lower
G1 E22 F300 ; purge nozzle quickly 
G92 E0 ; zero the extruded length
G1 E10 F30 ; purge nozzle slowly 
G92 E0 ; zero the extruded length
G1 E-4 F400; retract
G1 X100  F9000 ; pull away filament
G1 X150 Z0  F9000 ; pull away filament
G1 X180  F9000 ; wipe
G1 Y20  F9000 ; wipe
G92 E0 ; zero the extruded length
G1 E3.5 ; feed filament back
G92 E-0.5 ; zero the extruded length

Sorry that's a big post!

[Bking1340]

Are these S3D support team from china - don`t seem like they understand any english
Email to and from S3D support:
Read from bottom
--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------

Hi Jaco,

I'm glad to hear everything's in working order now. Please feel free to email me back if you have any other questions. Thank you again for contacting Simplify3D support.

Regards,


James Wagner | Product Support
Simplify3D All-in-One 3D Printing Software
Website | Facebook | Twitter | YouTube

On Tue, Dec 15, 2015 at 1:45 PM, Jaco Engelbrecht <jacoe79@telkomsa.net> wrote:
No also checked my G-code a few days ago, but did not see any ill effects.

From: James Wagner [mailto:james@simplify3d.com]
Sent: 15 December 2015 07:42 PM
To: Jaco Engelbrecht; Simplify3D Support

Subject: Re: S3D - Contact - Technical Support - Jaco

Hi Jaco,

Thank you for your reply, and yes I've seen that there are other users with similar issues. The file you've sent me does not display any retraction issues when I generated G-Code from it. Does it cause issues from your installation? In the G-Code the retraction values never are defined as anything higher than 1.5mm and when there is retraction during the wipe movement the values for retraction never add up to more than 1.5mm.

Are you seeing different results in your G-Code than what I'm seeing on my end?

Thank you,


James Wagner | Product Support
Simplify3D All-in-One 3D Printing Software
Website | Facebook | Twitter | YouTube

On Tue, Dec 15, 2015 at 12:12 PM, Jaco Engelbrecht <jacoe79@telkomsa.net> wrote:
Okay 1st off - do you see at the following links that I`m not the only person with this problem? : viewtopic.php?f=9&t=3536&p=15766&hilit= ... ent#p15766 and viewtopic.php?f=9&t=4066&p=17738&hilit= ... ion#p17738

The problem - When printing on Simplify3D 3.0.2 with "Perform retraction during wipe movement" selected, the retraction are sometimes 1.5mm as per my settings(Process 1 FFF) and sometimes it`s +-5 to 7mm which are nowhere in my settings.

Then - It`s not only one print that I`m having problems with, it`s 95% of my prints (different stl files). Have a original E3D V6 Hotend and standard extruder motor.
If I print anything on Simplify3d I have this problem.
If I print anything on Repetier Host(Slic3r), I DON`T have this problem
So mechanically my printer is 100%

Attached is my Factory File

Thanks

Jaco


From: James Wagner [mailto:james@simplify3d.com]
Sent: 15 December 2015 06:29 PM
To: jacoe79@telkomsa.net
Cc: Simplify3D Support
Subject: Re: S3D - Contact - Technical Support - Jaco

Greetings Jaco,

Thank you for contacting Simplify3D support. I'd be happy to assist you with this. Can you send me a factory file for the print you're having issues with?

If you are not familiar with a Factory File, it is the "project" file within Simplify3D software. It will save any 3D models that are imported, process settings, placement information, support structures - basically everything I need to re-create exactly what you were seeing but on my computer so that I can work from there. To save a Factory File, setup the program like you did previously, by importing the 3D models, configuring your Process Settings, and right before you would normally click Prepare to Print!, go to File--> Save Factory file As.


Please attach this .Factory file in an email back to me, and I'd be happy to start troubleshooting. Thank you again for contacting Simplify3D support. I look forward to hearing back from you.

Regards,


James Wagner | Product Support
Simplify3D All-in-One 3D Printing Software
Website | Facebook | Twitter | YouTube

On Sun, Dec 13, 2015 at 9:29 AM, no-reply@simplify3d.com <no-reply@simplify3d.com> wrote:
From:
Jaco - jacoe79@telkomsa.net

Short Description:
Simplify3D 3.0.1

Issue Details:
Good day
Can you please provide me with a download link or tell me how to downgrade from 3.0.2 to 3.0.1 as I have the same problem as these guys: viewtopic.php?f=9&t=3536&p=15766&hilit= ... ent#p15766

This is the thread that I have opened: viewtopic.php?f=9&t=4066&p=17738&hilit= ... ion#p17738

Operating System:
Windows

Printer:
Prusa I3

Attachment:
-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
So he asked me to see if my G-code file show any different results.
I Told him no - don`t see any ill effects
He think my problem is resolved because I don`t see any problems in my g-code

I don`t know what to say