I’ve got my k8200 for 1 weeks now, after finishing the build my print quality was terrible.
after installing a new support for the Z axis motor, a “anti-wooble” Z axis fixation, a fan hose, a sanded 2mm glass for the bed,marvin V2, etc…
i finaly got a better print quality among the Z axis
but the parts i made are no enough nice/precise for what i want to do (build small robot part)
after the big z problem, i’m looking to increase the precision to avoid to manualy reworking the pieces to fit each other)
As far, i didn’t find a “newbee friendly” tutorial to calibrate Slic3r, so i’m askin you. (if someone has one that tell/show “if you got this , change this”, i’m calling him “my god” for 1 year )
here is where i stand and the problems i’m facing (printing PLA @ 175°):
there are some blob along surface ( i’ve already check/measure the filament et the feed lengtt, it’s correct, retraction is set to 5mm)
it seems to me that it’s related to 1) : edge of hole and rounded part are realy messy.
As i’ve go a lot of “string” when the head travel over gaps ( and the retraction seems not to work at 190°) i reduced the temp of the hotend.
Changing from 190° to 175°, seem to almost solve the “strings problem”, but now i’ve got extrusion problem ( message “cold extrusion…”) as temp seems to vary over time ( by 5/7° somethimes).
i can change the marvin trigger value for this, but i assume it’s not the best way to go.
The surface are not realy regular. (but if i use calibration file 0.5mm-thin-wall.stl the wall is almost perfect, except for the problem 1) at one corner)
Fine tuning your slicer is mostly a trial and error challenge.
From your pictures it looks you have too much filament extruded hence the blobs and over-sized layers. 175° is too low a temperature to get good results. When you get the “cold extrusion prevented” message there’s no filament extruded explaining the gaps in some of your layers. Melting point of the filament varies a bit with the brand and color (black and yellow are oozing more than white and orange).
Speed is indeed very important. Try to print a small object putting manually the speed at 50%. It should improve the quality (I’m using low speeds between 30 and 50 mm/s).
Extrusion width is another important parameter.
To help you a bit further you should show us you current Slic3r settings.
the bed is exactly at 0.21mm (bristol paper size) and perfecly flat (2mm sanded glass)
i’ve already calibrate the extruder steps : i’ve mesuring the extrusion as described in the videos, the length vas exactly 100mm and my filament average width is 2.96mm, calculation give me the exact same value as in marvin V2
at 190°, quality is lot worse : I forgot to say blobs appear espetialy when the head travel over gaps.
i’ve reduced temp from 190° to 175 (by step of 5°) because at 190° there where more stringing and more blob (retraction seems not to work at 190° and at 175° i’ve the “cold extrusion” message when the temps drop by 5°).
my slicer setting are the “vellement_pla_new.ini”, except for the temp of hotend (175) and the bed temp (60°).
according to the reprap wiki, my layer heigth is correct (test with the stl of a 2cm square of 0.5 width give me a perfect straight face)
you said all info are on the web but, some say the opposite to each other it’s, for exemple here we said 175 is too low, on the reprap wiki they say PLA is 175°.
About the temperature: if you actually need lower temperature settings, you have to tell Marlin that those values are ok, so you need to reload the firmware. But first, check that the temperatures are correct, e.g. that the thermistor is in place inside the hole, and that there’s low airflow at that side of the heater block (otherwise, use e.g. a bit if Kapton tape for insulation).
PID autotuning should not be necessary here, only if you replace the heater cartridge or change to a different hotend. Monitor whether the temperature fluctuations happen when the fan turns on or off, that’s the most likely cause apart from a thermistor mounting problem.
In the above pictures, it looks like you may still have some Z wobble. Did you check if the distance of the “waves” matches the pitch of your Z rod?
I’ve try other calibration files with my temp reseted to 190°.
my problem seems to be bridging and oozebaning
the problem seems to be on bridging (here a photo before the 3 last layers)
i’ve try to change the “bridge flow ratio” but no amelioration
and there is also a problem of oozebane
i’ve try to change
retraction length to 5mm
speed to 250
extra length on restart to 0.05 instead of 0.1
no luck.
it seems to me that the filament is too liquid when it flow from the hotend.
It can’t retract well and bend a lot when i try to bridge. (that’ why i’ve try lower temps to 175°.)
any other suggestions ?
ps: as you see i don’t use capton to isolate the thermistor, but i use teflon ribbon. (i don’t have capton). and i use this fan hose http://www.thingiverse.com/thing:174247 ( a little bit cut because it touch my hotend)
Teflon tape is ok, just remember that you should not exceed 250 °C hotend temperature then.
How did you change Bridge flow ratio - up or down? Cooling enabled for bridging? Which speed and acceleration?
As for the retract settings: 250 mm/s doesn’t have any effect, unless you have modified the extruder a lot. With the original hardware (and E steps setting), 65 mm/s is pretty much maxed out. Should be audible that there’s no notable speed gain for higher values. You can set Extra length on restart to 0, it’s usually not necessary.
When slicing the single wall, what perimeter width is given in the G-code’s comment section? Looks like you have a temperature problem there, but could also be a wrong extrusion multiplier.
i’m not at home, so i can’t give you the exact numbers of things i didn’t change.
i’ve try 0.3 up and down (by step of 0.1) for the bridge flow ratio with no success. Cooling, speed an accel are the default from “pla vellement new”
for the extruder, i’ve extrude 100mm, and there were exactly 100mm extruded, my filament is a velleman witch has 2.96 average width so the value in marvin must be correct. (computation give me the same number as set in marvin)
for the extrusion multiplier i’ve change the default 1.05 to 1. At 1, i have tiny holes on top and bottom layers. I try 1.03 … less holes, so i came back to the original value of 1.05.
for the retraction, i’ve already try 0 for “extra length on restart” with no change.
I’m not already fluent with the GCode, i don’t know for the perimeter width , wich command must i search ?
I also think it’s a temp problem (as far as a newbee can think ), the best result was at 175°, but with the temp variation accros the print, the temp sometime decrease to less than 170° witch is the trigger for the “cold extrusion” securiy in marvin. I can change the value in marvin, but i don’t think it’s the best idea.
ps: i’ve check the thermistor, ohm are correct and @roomtemp it’s the same as the bed (and the room of course ). So i don’t think there’s a problem there.
When you slice something in Repetier you have generated G code on G-code editor tab and in the top you have commented out options that slicer used to slice your .stl file.
If you check line number 14 it should say what extrusion width to expect.
G21 ; set units to millimeters
M107
M190 S60 ; wait for bed temperature to be reached
G28 ; home all axes
G90 ; use absolute coordinates
G21 ; set units to millimeters
G92 E0
M82 ; use absolute distances for extrusion
M109 S190 ; wait for temperature to be reached
G1 E10 F400 ; prime the nozzle
G90 ; use absolute coordinates
G92 E0
M82 ; use absolute distances for extrusion
G1 F15000.000 E-5.00000
G92 E0
G1 Z0.300 F12000.000
G1 X89.650 Y79.650 F12000.000
G1 E5.05000 F15000.000
G1 X90.800 Y78.710 E5.07582 F1530.000
G1 X92.110 Y78.020 E5.10155
…[/code]
as you can see, the perimeter extrusion width is 0.38 mm in your case. If you measured a wall thickness of 0.5 mm on the thin wall print, your extrusion multiplier (or E steps) may be wrong.
You said that you calibrated the E steps - did you use a low extrusion speed, similar to the speed of the gears during normal printing? Doing the calibration too fast may cause slippage between filament and hobbed bolt, resulting in low filament feed and consequently too high E steps calculation.
In order to correctly slice such a thin wall model, manually set the extrusion width (Print Settings - Advanced) to the same value that’s used in the file. Or, second option, slice a solid model with 1 perimeter, 0 infill and 0 top solid layers - this will also produce G-code like the thin wall model, but without worrying about whether the widths in Slic3r and the model are the same.
ps: for extrusion i use the repelier interface. i demand 100mm extrusion, and i get exactly 100mm extruded, so i assume e-step is correct. the filament is velleman and get a average of 2.96mm.
Just one thing about that thin wall print: try to measure as little layers as possible. When you measure all layers like in your photo, you will measure a combination of the wall thickness and some print errors like (low, but present) wobbling, filament irregularities etc.
About the E steps setting at 100 mm extrusion: your method is correct. Just keep in mind, that the speed of the test extrusion is important. More speed equals more pressure in the hotend, thus more necessary force to push the filament. At some point, the hobbed bolt looses grip and the filament starts to slip, which of course reduces the extruded length. I.e. when you do this test too fast, it’ll seem like your E steps are too low.
One more thing: when you try the hollow cube or a similar part next time, lower the bed temperature to 50 °C, to reduce the softening of the bridging PLA due to heat from the bed. Maybe you can even go lower than 50 °C for the upper layers, depending on how well your PLA sticks to the surface.
quick littles questions as i try to check if the thermistor give correct high temperature value. (lower temp is correct and resistance is same as the construction manual said)
At wich temp filament start to come out of the hotend ? ( for me its between 150 and 170° when i start heating the hotend)
Is it normal when the print is finished that the PLA continue to come out of the hotend so much ? Almost 15 cm of 0.5 thin filament when cooldown from 190°C.
Could it be that temp measurement is incorrect? Perhaps the actual temp is higher than the reading.
What is the temp you read when the hot end is cold?
Have you tried to measure the hot end temp with an external thermometer/ e.g. IR thermometer?
Same for me - filament starts to flow by itself at about 160
That seems a lot to me, but it does vary with what colour you’re using. I’ve found that natural runs out more than black, so I’ve taken to printing black at 185 and natural at 180. With natural at 180 I get about 1.5 cm exuded when the print finishes, black at 185 gives about 1 cm.
)
