Crowning bug when continuous width change going through a radius in Z.?

58 views
Skip to first unread message

Jeremy P

unread,
Sep 12, 2014, 11:29:21 AM9/12/14
to kisslicer-r...@googlegroups.com
Hi, 

This is going up a radius in Z toward the inside. You can see the support on the inside getting ready for it.
The old KISS is pretty good about keeping a full crown pass in between the normal passes.

The newest version ends up getting really sparse and many passes lack it entirely for about 10 layers and  de-laminates or it never laminates at all and it kills the part.
One side of the part was worse than the other, and it never stuck on the one side.
It looks sort of like the picture for about 10 layers before it finally pops another full pass in there.

It looks like it happened around the Z height where each move up in Z is a rather large move in radially and it just put the filament in a hollow spot that it created.

The old KISS was able to keep varying the rest of the perimeters while keeping a crown in there the whole time.

 



width calculations.jpg

lonesock (Jonathan)

unread,
Sep 12, 2014, 12:18:30 PM9/12/14
to kisslicer-r...@googlegroups.com
Hi!.

Could you please post the STL and let me know your extrusion width, oversample, and slice thickness settings?

thanks!
Jonathan

Jeremy P

unread,
Sep 12, 2014, 12:58:39 PM9/12/14
to kisslicer-r...@googlegroups.com
Jonathan,

My extrusion witdh is .8, layer is .2 mm ,  over sample was .25 at the time and give about 8 -10 layers of sparce to nothing, and at .125 sample it does it for about 5-6 layers.
Crown threshold at .001mm :),

I guess under normal circumstances it should have been ok, but because the top part was .125 in thick, it pulled hard enough to delaminate it.

Part of it is operator error I am sure, but I wonder if there is a way to get KISS to do solid loops regardless? If in doubt, do a solid loop?   lol
Though it does seem a little different than the old KISS where I think it was only about 1 layer to transition to two close together  blue colored loops again after getting away from the crown pass on the previous layer.


Jeremy


CAP.STL

Jeremy P

unread,
Sep 12, 2014, 4:34:40 PM9/12/14
to kisslicer-r...@googlegroups.com
I am trying to run just the top part of that file(last half inch of the part I sent you) , and i get about .2 inches up and it looks like I loose steps in one axis and the whole print gets shifted over.  I am wondering now if it was not a problem with  delamination, due to lack of crowning, but the printer slamming into curled up support.  I think when KISS trys to start making the smaller support before laying actual part on top of it, it doesnt connect all of the ends to the existing course support so they just curl up in mid air and start to get int he way of the printer and it slams into them and skips on the stepper.

If KISS would try to stay as close to 90deg of the course support when laying on top of it for the actual part material, and it made sure to always bridge to the next course support fin(rather than stopping in between them), maybe it would be stuck down and stay at the z height it should, rather than curling up. 

I tryed the z move up after wipe, but KISS makes many rapid moves across the part that do not have a wipe, like while doing support material.
I made the support finer, and I am going to try to watch as much as I can to catch it in action. hmmm

Jeremy P

unread,
Sep 14, 2014, 4:02:41 PM9/14/14
to kisslicer-r...@googlegroups.com
Ok, so I just made my support type  "fine" and it was close enough that when it put the grid on top of it to begin laying part passes on the support, it bridged and connected to most of it.
This way it didnt just curl up and jam the printer.  I guess this is sort of an obvious approach, but it kind of seems the course settings are not really useful then.
 The problem is, some of the parts are 10 inches in diameter lol. and sort of tall. It makes a difference of like 200-300 cm3 ish, not to mention many hours more.

Just wish KISS payed attention to where the course support is and made sure to stick the end of the extrusion on the next closest support fin so it would stay out of the way and not curl up and harden.

On another great note,  AFAIK, KISS is the best slicer out there. I just visited cura and slic3r to see what they would do in this situation, and I would not even be able to make a part with that junk.
In fact, I have never been able to make anything without Kisslicer.

Jonathan, you definitely know how to lay filament dude. lol  You are apparently a freaking ninja too. I bet you have those split toe ninja shoes on when you are slammin on the keyboard.

btw, after trying those other two slicers again, I am so freaking happy to have Kisslicer I am going to start bribing you for features lol.

How much do you want to add a control box so I can set my outer perimeter speed to its own adjustable speed?
 IE, run it slower to get more accuracy in an out of a bunch of sharp corners heh.
Reply all
Reply to author
Forward
0 new messages