Bug in retraction / layer change

lohiaprateek
Posts: 7
Joined: 15 Feb 2019, 06:27

Bug in retraction / layer change

Post by lohiaprateek »

Ive been a long time user of Kisslicer and possibly started using it 2012/13 when I got my first printer. And have always found it to be consistent in the code output, but I've encountered a bug in the retraction / Preload setting in v2 and 1.6

Even when I set preload to 0 and z-lift to 0 the printer forces an arbitrary z-lift on every move which should have a preload retraction, it does this even on layer change, with preload set to 0 this z-lift and drop makes for an extremely ugly seam as filament forms a blob whenever it lifts and and fall down.

This lifting and dropping is also adding layer lines on the prints, which are not present on prints from other slicers.

I am attaching a gcode file.

An example can be found on line 57777 and 57778 of the gcode file. The code shows that it lifting the head by .56 mm and subsequently dropping it in the perimeter loop, causing a blog.
Attachments
cone.zip
(380.45 KiB) Downloaded 79 times
User avatar
pjr
Posts: 692
Joined: 05 May 2015, 10:27
Location: Kamnik, Slovenia

Re: Bug in retraction / layer change

Post by pjr »

The Z-lift is present because the minimum layer time (5 seconds in your settings) has not been reached.

The extrusion is slowed for the layer, but there is still a delay required prior to the next layer, and this delay requires a z-lift.

You could reduce your layer time or minimum volumetric speed or both.

Peter
lohiaprateek
Posts: 7
Joined: 15 Feb 2019, 06:27

Re: Bug in retraction / layer change

Post by lohiaprateek »

Have reduced min flow to 0.01 and changed minimum layer time to 0, but there is still a variation of 0.5652, in lines 54641 and 54642. I am attaching gcode file and also the STL in one zip file.
Attachments
cone1.zip
(374.92 KiB) Downloaded 79 times
User avatar
pjr
Posts: 692
Joined: 05 May 2015, 10:27
Location: Kamnik, Slovenia

Re: Bug in retraction / layer change

Post by pjr »

OK, I see that now.

I will try to find out if that's a fault or by design.

Peter
lohiaprateek
Posts: 7
Joined: 15 Feb 2019, 06:27

Re: Bug in retraction / layer change

Post by lohiaprateek »

Thank you !!
User avatar
pjr
Posts: 692
Joined: 05 May 2015, 10:27
Location: Kamnik, Slovenia

Re: Bug in retraction / layer change

Post by pjr »

OK, so it's not a fault...

You have a Z shrink value of 1.3%; therefore Z will be printed 1.3% higher. There is no Z-lift and lower in this G-code.

The layer to which you refer, I think is this:

Code: Select all

; time estimate: pre = 2.915026, post = 2.206763
; Dwell time remaining = -2.206763
;
; Post-layer lift
G1 X3.816 Y-0.95 Z43.1611 E0 F12000
; END_LAYER_OBJECT z=42.350
;
; *** Every 1 Layers ***
M117 169
;
; BEGIN_LAYER_OBJECT z=42.600 z_thickness=0.250
;
; 'Loop Path', 2.0 [feed mm/s], 50.0 [head mm/s]
G1 X-0.069 Y3.89 Z43.1611 E0 F12000
G1 X-0.069 Y3.89 E0
; head speed 49.999996, filament speed 2.016396, preload 0.000000
G1 X-0.177 Y3.875 E0.00431 F3000
A Z of 42.6 will be printed at 1.3% more = 43.1538. Although this is not precisely the print value, I think it's close enough.

Please let me know if you think I am wrong (It does happen quite a lot :( )

Peter
lohiaprateek
Posts: 7
Joined: 15 Feb 2019, 06:27

Re: Bug in retraction / layer change

Post by lohiaprateek »

Yes, removing the z shrink does eliminate the random hops, very interesting, because I somehow felt that a shrink factor will actually scale the file uniformly in x, y and z automatically and remove the manual scaling of individual files depending on which material is used. I feflt this to be a rather benign operation. For parts with smooth surfaces the blob does become quite evident and that's when it came to my notice.

I started fine tuning my profiles a lot more over the past couple of weeks as I plan to start selling printers with a pre-tuned Kisslicer profile for industrial clients who have accuracy as their utmost priority, and thats when I calculated the shrinkage in the PET-G I use and plan to stock.

Although IMHO, I still think it is as a bug, as it severely worsens the print quality due to the blobs.
lohiaprateek
Posts: 7
Joined: 15 Feb 2019, 06:27

Re: Bug in retraction / layer change

Post by lohiaprateek »

And, I've encountered another retraction bug. The part has supports and infill, but when the head completes the infill and moves to the support, which is a new layer it does not retract the filament, but rather just retract as per the preload setting, even though the distance is greater than the trigger distance.

Kindly search for M117 Layer: 58, the code right before the layer change shows the preload suck but shows no destring suck. I guess the best possible way out would be to tune the preload settings more accurately.

You may download the gcode file from the link

https://1drv.ms/u/s!Aixf9Gp8GftHiscg8XlCxEld2ZMMmA
User avatar
pjr
Posts: 692
Joined: 05 May 2015, 10:27
Location: Kamnik, Slovenia

Re: Bug in retraction / layer change

Post by pjr »

lohiaprateek wrote:And, I've encountered another retraction bug. The part has supports and infill, but when the head completes the infill and moves to the support, which is a new layer it does not retract the filament, but rather just retract as per the preload setting, even though the distance is greater than the trigger distance.
This is a known fault in V2

Peter
Cthutq
Posts: 53
Joined: 12 Jul 2018, 13:45

Re: Bug in retraction / layer change

Post by Cthutq »

pjr wrote:
lohiaprateek wrote:And, I've encountered another retraction bug. The part has supports and infill, but when the head completes the infill and moves to the support, which is a new layer it does not retract the filament, but rather just retract as per the preload setting, even though the distance is greater than the trigger distance.
This is a known fault in V2

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