Bug in retraction / layer change

Re: Bug in retraction / layer change

Postby pjr » 17 Feb 2019, 08:44

Cthutq wrote:Peter, when you can expect to update V2 version, half a year has passed since the release of the alpha version


I wish I knew. Usual saying "It will be ready when it's ready". Sorry, we are all in the same position :(

Peter
Please note: I do not have any affiliation with KISSlicer. Any advices given are offered in good faith. It is your responsibility to ensure that by following my advice you do not suffer or cause injury, damage or loss.
User avatar
pjr
 
Posts: 562
Joined: 05 May 2015, 10:27
Location: Kamnik, Slovenia

Re: Bug in retraction / layer change

Postby lohiaprateek » 23 Mar 2019, 03:29

I've been trying to print https://www.thingiverse.com/thing:246198 also attaching the gcode. The problem is when it starts to print the feet, it does multiple preload adjustments within the feet which eventually ends up blocking the hotend. I then have to retract the entire filament, cut the end and then reprint, which causes it to block again.

It works perfectly when I use CURA 15.05 to print with the same retraction setting as in KISSlicer.

I use an original Prometheus 2 hotend with a e3d titan, recalibrated with the calibration tool by Michael Hackney from Sublime Layers, and I've got the error deviation down to .2mm over a 90mm extrusion. I even tried the same gcode with a Chinese J-Head made of PEEK and an aluminium heaterblock, which I know works perfectly and has multiple 100 hours of prints on and I even managed to jam that nozzle. The preloadVE tuning wizard cube with rounder corners prints perfectly, and I have managed to zero in on the prelaod value. On large prints with some filament extrusion between the preloadVE adjustments this problem does not manifest and it prints perfectly.

Maybe if there could be configurable threshold of .5 or 1 mm between any PreloadVE and retractions, this problem could be mitigated. Such a feature is present in the older builds of CURA, possibly to avert the same problem. Another possible solution could be to do a solid infill below a certain threshold, something similar can be found in Slic3r. This would also strengthen the part and also avoid unecessary retractions and preload adjustments when the infill is 10%.

Also file attached is in 7z format, regular zip file was too big to be attached.
Attachments
loubie_adalinda_dragon.7z
(9.97 MiB) Downloaded 15 times
lohiaprateek
 
Posts: 7
Joined: 15 Feb 2019, 06:27

Re: Bug in retraction / layer change

Postby pjr » 23 Mar 2019, 04:00

Well, Adalinda is very printable with the correct settings: viewtopic.php?f=15&t=2371&start=20#p8481

I will have a look at your G-code to see if I can see anything.

Peter
Please note: I do not have any affiliation with KISSlicer. Any advices given are offered in good faith. It is your responsibility to ensure that by following my advice you do not suffer or cause injury, damage or loss.
User avatar
pjr
 
Posts: 562
Joined: 05 May 2015, 10:27
Location: Kamnik, Slovenia

Re: Bug in retraction / layer change

Postby pjr » 23 Mar 2019, 04:46

lohiaprateek wrote:It works perfectly when I use CURA 15.05 to print with the same retraction setting as in KISSlicer.


I guess that if you turned off Preload in KISS then it would work as well.

When using Preload, you need a lot less retraction. E3D suggest retraction values of up to 1mm for the V6 to prevent blockages but you are not using direct drive, so 3mm would be about right for non-Preload slicing.

I suggest that you try setting retraction to less than 1mm and maybe also reducing the "Main" temps (maybe to 200). Printing at 215 degrees will usually encourage stringing. You are not printing particularly fast, so I don't really understand the relatively high temp.

You might also want to turn off wipe as well. And personally I would set the loop order to Loop1 > Perimeter.

Peter
Please note: I do not have any affiliation with KISSlicer. Any advices given are offered in good faith. It is your responsibility to ensure that by following my advice you do not suffer or cause injury, damage or loss.
User avatar
pjr
 
Posts: 562
Joined: 05 May 2015, 10:27
Location: Kamnik, Slovenia

Re: Bug in retraction / layer change

Postby lohiaprateek » 27 Mar 2019, 06:55

pjr wrote:
lohiaprateek wrote:It works perfectly when I use CURA 15.05 to print with the same retraction setting as in KISSlicer.


I guess that if you turned off Preload in KISS then it would work as well.

When using Preload, you need a lot less retraction. E3D suggest retraction values of up to 1mm for the V6 to prevent blockages but you are not using direct drive, so 3mm would be about right for non-Preload slicing.

I suggest that you try setting retraction to less than 1mm and maybe also reducing the "Main" temps (maybe to 200). Printing at 215 degrees will usually encourage stringing. You are not printing particularly fast, so I don't really understand the relatively high temp.

You might also want to turn off wipe as well. And personally I would set the loop order to Loop1 > Perimeter.

Peter


Thanks Peter, the jamming issue was solved by "seasoning" the hotend. I used some PLA which was tuned for PTFE lined direct drive hotends and not a full metal hotend (I know this because I spoke with the manufacturer), and somehow managed to wreck my nozzle. After multiple cold pulls to completely clean out the hotend and then seasoning it, I've now got reliable PLA extrusion. I have also setup a small filament oiler to keep things nice and clean.
Printing was at 215 because the material had a 210-230 sticker on it. Hence I chose 215.

I now have a new PLA supplier and the new material prints reliably at 205 with .3 mm laters and 60mm/s extrusion speeds, with a custom designed and printed direct drive extruder. My preload is now down to .68 while retraction remains at 1.8. I have printed multiple tuning wizards for both preload and destring, and even tuned it by the eye, a preload higher than leads to filament starvation, where extrusion starts are blanks for a couple of mm and then begin to extrude, and destring any lesser gives me strings.


I havnt printed Adalinda after changing materials, but some other prints are printing cleaner
lohiaprateek
 
Posts: 7
Joined: 15 Feb 2019, 06:27

Previous

Return to Bug reporting

Who is online

Users browsing this forum: No registered users and 1 guest

cron