Multiple bugs making S3D pointless for dual extrusion

[difool78]

Hi

I am printing dual, either for soluble supports or dual colors. I've chosen S3D in the past for that capability and better quality results.

I've done some dual colors parts with version 3.1, it was a no brainer. I have upgraded to 3.1.1 when released.
Since two weeks, I've restarted printing dual mainly for supports, but also for a dual color part. I have a customer who ordered a part in two colors and others needing dissolvable supports, as high quality is required.

Printing these parts with S3D 3.1.1 is a PITA due to multiple bugs around additions and settings for dual printing.

1) I have reported a first bug earlier this week on the absence of retraction between ooze shield and part
viewtopic.php?f=9&t=6420
workaround disabling "only retract when crossing open spaces" is creating blobs on the outer perimeter, degrading the print quality.

2) There is also the same bug between prime pilar and part, no retraction at all.
So adding these two issues when printing dual with prime pillar and ooze shield, is the idle hotend is spilling material all other the print bed and creating gaps on the part.

3) There is a bug using brim and ooze shield : brim is not on the part, but on the ooze shield
So bed adhesion is compromised

Brim issue

4) There is a bug when adding a tool change script. This creating an offset, at each toll change, of some 1/10th of a mm in X and Y so the print, ooze shield and the prime pillar are leaning. Removing the scripts, no more leaning.

Leaning parts

Mat - copie.factory

(97.91 KiB) Downloaded 331 times

It seems also that the same effect appears when only printing prime pillar, without ooze shield and without any tool change script.

Given all these bugs, it is now pointless to print any dual extrusion file with S3D 3.1.1

So I will have to revert back to Cura at least for my dual extrusion parts. Seems that my made improvements for the Ultimaker 3.

Any feedback on these bugs from the support team will be appreciated.

Regards
David

[arhi]

2) There is also the same bug between prime pilar and part, no retraction at all.
So adding these two issues when printing dual with prime pillar and ooze shield, is the idle hotend is spilling material all other the print bed and creating gaps on the part.

I'm also having a lot of issues with s3d and dual extruders but I'm not sure I understand this problem properly? why do you expect retraction between prime pillar and part? isn't the whole point of prime pillar to "fill in the nozzle" so that there's no issue on part? Or the prime nozzle is just to prime the stuff that "oozed" from the extruder and you expect regular retraction leaving primer and going on to part .. hm .. interesting .. I never had issues with strings on the prime pillar as it's far away from the print..

[brian442]

For #3, I would actually bet that's intentional, since I know that I have FAR more trouble getting my ooze shield to stick to the bed than I do with the part. And typically the ooze shield is positioned as close as possible to the part (usually only a millimeter or so away), so it already helps to hold the part down. I for one would be quite upset if I couldn't use the brim to hold down the ooze shield since that detaches far more easily than the part does.

For #4, if it's a bug, it's probably in the firmware not the software. Even when I slice and look at the gcode file, all of the XY coordinates stay correct. They are not shifted. So if the print itself it moving to different XY locations, then maybe it has a bug with roundoff error or something similar. Again, you can look through the gcode file to verify the S3D commands are correct. One of my old firmwares had this exact same bug where the prints would continually lean in one direction, and it was eventually found and patched. So firmware bug seems more likely since software commands are correct.

[difool78]

2) There is also the same bug between prime pilar and part, no retraction at all.
So adding these two issues when printing dual with prime pillar and ooze shield, is the idle hotend is spilling material all other the print bed and creating gaps on the part.

I'm also having a lot of issues with s3d and dual extruders but I'm not sure I understand this problem properly? why do you expect retraction between prime pillar and part? isn't the whole point of prime pillar to "fill in the nozzle" so that there's no issue on part? Or the prime nozzle is just to prime the stuff that "oozed" from the extruder and you expect regular retraction leaving primer and going on to part .. hm .. interesting .. I never had issues with strings on the prime pillar as it's far away from the print..

Having no retraction between prime pillar and part allow oozing, specially with PVA and with Bowden extruders.
In that configuration, even with retraction you have oozing when idle at Printing temp. So prime pillar is used to replace this stuff who as leaked from the nozzle, then I need a retraction to stop oozing the time to move across the part and print the support. Prime pillar and supports can be quite far each other.

[difool78]

For #3, I would actually bet that's intentional, since I know that I have FAR more trouble getting my ooze shield to stick to the bed than I do with the part. And typically the ooze shield is positioned as close as possible to the part (usually only a millimeter or so away), so it already helps to hold the part down. Anyways, I for one would be quite upset if I could use the brim to hold down the ooze shield since that detaches far more easily than the part does..

Purpose of the brim is to improve adhesion of the part, so with or without ooze shield it should hold the part on the bed.
Agreed that ooze shield adhesion is tricky except printing two perimeters or more, but if it détaches during print impact might be lower than if it's the part itself.
Ooze shield is quite thin and breakable, so I would suggest that when having brim + ooze shield, part and ooze shield are hold down by the brim.

[difool78]

For #4, if it's a bug, it's probably in the firmware not the software. Even when I slice and look at the gcode file, all of the XY coordinates stay correct. They are not shifted. So if the print itself it moving to different XY locations, then maybe it has a bug with roundoff error or something similar. Again, you can look through the gcode file to verify the S3D commands are correct. One of my old firmwares had this exact same bug where the prints would continually lean in one direction, and it was eventually found and patched. So firmware bug seems more likely since software commands are correct.

I have made a mistake in the first post : the problem is when I have prime pillar without ooze shield but tool change script may have an impact.

The point is that using another slicer, Cura, I do not have the issue. Printing a part with ooze shield + prime pillar with S3D 3.1.1 is correct.
The issue is when only printing the prime pillar or the ooze shield only
I will have a look at the gcode using prime pillar, ooze shield, both and also have a try with previous version of S3D before considering a firmware issue (which hasn't changed between the good prints and today)

[difool78]

For #4, if it's a bug, it's probably in the firmware not the software. Even when I slice and look at the gcode file, all of the XY coordinates stay correct. They are not shifted. So if the print itself it moving to different XY locations, then maybe it has a bug with roundoff error or something similar. Again, you can look through the gcode file to verify the S3D commands are correct. One of my old firmwares had this exact same bug where the prints would continually lean in one direction, and it was eventually found and patched. So firmware bug seems more likely since software commands are correct.

I have made a mistake in the first post : the problem is when I have prime pillar without ooze shield but tool change script may have an impact.

The point is that using another slicer, Cura, I do not have the issue. Printing a part with ooze shield + prime pillar with S3D 3.1.1 is correct.
The issue is when only printing the prime pillar or the ooze shield only
I will have a look at the gcode using prime pillar, ooze shield, both and also have a try with previous version of S3D before considering a firmware issue (which hasn't changed between the good prints and today)

I have checked the code generated by S3D 3.0.2; 3.1.0 and 3.1.1 on the prime pillar, as it the same layer after layer. And it is always at the same position. So the problem is not on S3D side for the #4. So it seems it's more on firmware side, even if Cura is not creating the issue.
BTW Brina 442 do you have any clue on where was the bug in the firmware ? Was is a Marlin firmware ?
Thanks in advance
David

[difool78]

2) There is also the same bug between prime pilar and part, no retraction at all.
So adding these two issues when printing dual with prime pillar and ooze shield, is the idle hotend is spilling material all other the print bed and creating gaps on the part.

I'm also having a lot of issues with s3d and dual extruders but I'm not sure I understand this problem properly? why do you expect retraction between prime pillar and part? isn't the whole point of prime pillar to "fill in the nozzle" so that there's no issue on part? Or the prime nozzle is just to prime the stuff that "oozed" from the extruder and you expect regular retraction leaving primer and going on to part .. hm .. interesting .. I never had issues with strings on the prime pillar as it's far away from the print..

Having no retraction between prime pillar and part allow oozing, specially with PVA and with Bowden extruders.
In that configuration, even with retraction you have oozing when idle at Printing temp. So prime pillar is used to replace this stuff who as leaked from the nozzle, then I need a retraction to stop oozing the time to move across the part and print the support. Prime pillar and supports can be quite far each other.

I had a closer look at the GCode on S3D 3.1.1, 3.1.0 and 3.0.2 and there is a retraction after prime pillar, but if there is a tool change just after the prime pillar, it's the retraction set in tool change retraction which is applied, not the one set in the extruder layer.
In my case, having E3D v6 direct hotend I need to limit retraction to avoid clogging the nozzle. I have applied the minimum commun denominator, though with PVA I am able to increase the retraction. Thats why I had oozing with PVA after the prime tower.

My setting was to use prime pillar only with right extruder (with PVA) (and ooze shield only with left extruder), so when not printing supports, I was doing the prime tower with right extruder, then tool change and part printing. I've just reverted to use both extruders for prime pillar and ooze shield to avoid oozing when not using the nozzle and also decreasing the print time.

[Zorgonaute84]

We already talked about that, but +100 about the missing retraction between the prime pillar and the part.

Without this retraction youve got a lot of stringling between the prime pillar and the part, It's not too bad with dual extrusion PLA + PVA but it could very anoying with dual color...

Please S3D stop adding some bullshit features and start to fix all the issues of the dual extrusions process... (really, we don't care about translation... all forums are in english and it's better to keep the software in english when we start to learn in this language. I speak french, but I kept S3D in english for better understanding and visualisation...)