Cubex Infill and Filament Flow Failure

[scott.t...@scottscustomoffroad.com]

I've been using a Cubex Duo for about a year now. I use it as a prototyping tool and also to make usable parts for machinery. We started using Kisslicer and really like its functionality and its ability to allow the Cubex to use bulk filament. But, ever since we started using Kisslicer we've had filament flow failures, so much so that its almost unusable.

I have played with prime & suck functions in the g code, changed speeds, used a few different brands of filament, and tried many other adjustments, but nothing seems to solve the flow failure problem.

Today I started using Cubex firmware 1.07 along with KISSlicer_Win64_1_4_4_6 and CubeItMod 4.8. I was using 1.06 and thought 1.07 would help but it didn't.

The parts I've tried all look the same no matter what I change as far as the infill goes. When I'm watching the feed motor while its infilling, the feed motor continuously changes direction (prime/suck?) and does NOT advance (extrude) any further until it gets to the outer loops. This causes the infill to be VERY sparse, sometimes only a dot of PLA at the start and end of each segment of infill. It also seems that the head travel speed is faster than it is when extruding loops, and the speed settings I'm using are Loops: 25.2, Solid: 33.2, Sparse: 22.

The result from this is usually filament flow failure. I'm currently using Matterhackers blue PLA. I'm pretty sure the forward/reverse (prime/suck) action over a few seconds of time wears out the filament where the drive roller is contacting it, therefore causing it to slip and fail.

How can this prime/suck action be stopped and it feed filament normally so it doesn't fail during infilling?

[Peter Gregory]

What temperature is your first/additional layer set to?
Matterhacker filament has a narrow print range from 195 to 215.
If you print at low temps, it may not be able to extrude at the higher speeds needed for infill.
Do you hear a lot of clicking while it attempts to infill?

Have you loaded the CubeItMod language file?
There are several fields CubeItMod "hijacks" to give control over infill.
What is your infill percentage? If it is too low, the RPM will sit at .1 and very little filament will come out.

[giovanni...@gmail.com]

if possible post here a bfb and the related bak file.

[Scott]

The first layer is set at 230 and additional layers are 195. The temps are what was loaded with the new kisslicer.

I loaded the CubeltMod.po language file in. I don't know what hijacks there are yet, maybe that would help.

I have the infill at 25 percent. I notice when saving the file it runs Cubelt and in the Complete! window under Extruder1 it lists; M227 S118 P444, M228 S44 P44, Solid%=111  String X/Y speed multiplied by 0.9, Infill%=3  Infill X/Y speed multiplied by 33.33

Infill%=3, is that causing the rpm to sit at .1, if so, how do I change that?

[Scott]

Here's the bfb and bak files.

[Scott]

One other question/concern I have is that I'd like to have stacked infill like Cubex has. I thought kisslicer had the ability. I'm assuming that stacked infill means that all layers of infill go the same direction instead of alternating by 90 degrees, right?

[giovanni...@gmail.com]

set solid % and infill % to 100 like here :

the M227/228 S & P may be good but as I don't use Matterhackers filament I don't know.

Having all the layers go the same direction is not good at all because this adds constraint on each layer and the part warps more than with the crossed layers.

Stacked infill is used to increase speed, if you set Layer thickness 0.1 and stacked layers to 3, the printer prints external loops every 0.1mm but the internal infill will be printed only every 0.3mm.

[Peter Gregory]

The first layer is too hot for matterehackers PLA filament. Change it to 215.
The additional layers at at the low limit for matterehackers PLA. I'd up it to 200.
If it is too cold, it won't be able to push the PLA out.

[Scott]

OK, I did all those thing both of you said to try. They made a good improvement on the infill especially in the larger areas (.50"x.625"). But, in the smaller areas, .20"widex3.0"long it starts the infill on one side of the part and makes a dot. As it moves across the part at a 45 degree angle, it doesn't make a continuous strand from one side to the other, until it gets to the bigger area.

I increased the flow tweak to 1.19, main temp to 200, first layer to 215. Is there a setting to make it prime more on each strand of infill?  I have same setting as above; M227P=.05, M227S=.03, Wipe=.5, infill%=100, Solid%=100.

Thanks for the quick responses!

[Peter Gregory]

You will probably have to set the maximum printer speed lower.
Support infill runs really fast. It was too fast for my material to keep up (RPM would be 5.0, too fast)
I'm not at my machine to get my settings, try dropping the maximum print speed down.

[giovanni...@gmail.com]

OK, but now start the settings from the beginning.

Print the cyl30x10 with the layer thickness you want to tune, into "Style" set infill to vase, extrusion width 0.5, Skin tickness 0, number of loops 1, infill extrusion width 0.5, seam hiding 0.5, jitter 0, Wipe ON, Use M227 ON.

On "Support" Prime pillar None

Print and measure the perimeter, if it's 0.5mm and if the perimeter is without small holes your Flow tweak is OK, otherwise you have to play with Flowtweak until the perimeter is 0.5 an it looks good.

When the flow tweak is OK you can find the settings for the M227 & 228, use the same settings like before , just change the number of loops to 2 and loops go from inside to permieter ON.

Print and look at the seam to adjust the M227&228 parameters.

Once your seam looks good you can set the Solid% and the Infill% by printing the TestSolidCourt.

Go to "style" set Infill to 33%, skin thickness 1, number of loops 2 or 1, stacked layers 1 to 4, print.

The longer paths on Solid and Infill should be good as you set the Flow tweak before, you may adjust for short pats with Infill% and Solid%.

If you notice a lack of material on short paths of Infill and/or Solid increase Infill% and Solid% ( not too much just something like 110 or 115 should be OK ), if you have too much material on short paths decrease a little ( 85-95).

Once you did this your settings are good and you just may have to do some little adjustments to fit the shape of the parts you want to print, but you have a good setting to start from.

[Scott]

What does M227 S&P and M228 S&P do???  Anything I change on them makes the print worse. The "TestSolidCourt" worked great. The infill looked the way it should because it was one continuous extrusion. On my part attached, the infill is made of small segments, in between which it does a prime/suck/(maybe wipe, not sure). These small segments seem to cause the problem. If there was just one prime/suck/wipe at the beginning and end, and the infill was continuous, I think it would work great. This also causes the loop after the infill to be thin and not have enough plastic, like there's too much suck and not enough prime.  

[giovanni...@gmail.com]

Your problem is caused by M227/228 parameters, read my post of 17 june here :

https://groups.google.com/forum/?hl=fr#!topic/kisslicer-refugee-camp/VvmF70oNQEI%5B101-125-false%5D

it would help understad how these parameters work, when you find the right setting the seam is very good.

The M227/228 parameters depends essentially  on the filament you're using, the temperature you're working and the layer thickenss, but also other parameters like the speed have some influence. 

Look here, the seam is on the middle of the back.

Here are my settings for another filament's brand with layer thickness 0.2, you can try them :