And another example… Left is the best I could get with a simple 20mm cube and slic3r playing with it for weeks. If i increased the extrusion factor it would print out of calibration, and after recalibrating I’d be back to the gaps. Right is with SFact…
I second the Cura recommendation. Works nicely under Linux.
@Rob_Antonishen Wow - any difference in temperature / speed that you know of?
I find that in Slic3r you MUST specify the “extrusion width”… if you leave it as default, god knows what it will use…
I set the extrusion width to my nozzle diameter as a starting point, and can increase it from there up to around 250% for different parts… You don’t want to extruder less than the nozzle diameter, and often you will only want 100% (of your nozzle diameter) for the perimeters, with thicker infill.
It looks like those threads on the left part are very sparse, which slic3r does seem to do when the extrusion width is left at default…
@Jarred_Baines The temps and speeds are the same. I agree it is an issue with the extrusion width settings. I calibrate thin wall perfectly, with the thickness exactly what the gcode comment states, and this is the result. It seems to be an issue with J-Heads based on looking in the forums…
Please post results, I too have a jhead and have similar problems. However increasing my nozzle diameter seems to help.
I select the slicer on a print-by-print basis. I have achieved complete infill like the right example here using vanilla Skeinforge. For some prints Slic3r gives a more complete fill. I have not tried Cura yet.
RepRap mag. 1 has a good overview of slicers that might come in handy if you want to choose the right slicer for the job.
http://reprapmagazine.com/issues/1/index.html
Might be worth a read.
rob you need to remove your hotend and calibrate your e-steps properly. measure at 10CM+ to reduce error. I would not be surprised to find out that your calibration is wrong.
@James_Zatopa thanks for the suggestion, but I’ve already calibrated the extruder esteps and double-checked it. I recalibrate esteps for every different filament I use as it will vary slightly based on the rigidity and depth of bite the hobbed bolt takes.
The results are 2 difference slicers…
You can’t say the esteps are wrong for the slic3r one and correct with the sfact one…
If the machine hasn’t been re-calibrated between the prints, you can only say it’s to do with the slicer (unless another problem mysteriously crept in between prints, extruder playing up or something similar)
@Rob_Antonishen that is way more extreme then even I take it and I am quite picky. I calibrate for any filament then measure the filament I am printing with and input that into the slicer I am using. Are you measuring your filament and using that measurement in your slicer? I prefer kisslicer but I do use Slic3r and have not had the issues you are having.
For those of you that don’t know measure both x and y of filament and use either an average or the smaller of the two measurements
@James_Zatopa
http://reprap.org/wiki/Volumetric_Dimension_settings
The first few lines and the first diagram on this page describe why you need different e-steps for different filaments, it’s not so much about the DIAMETER of the filament, nor the outside diameter of the hob, it’s about the “Drive diameter” (probably not the correct term).
PLA is hard and so the teeth don’t bite it as much, so it gets driven (let’s assume an M8 hobbed bolt) at around the outside diameter of the hobbing…
ABS is softer, so the teeth dig into it more, and it reduces the effective diameter of the drive, instead of being driven by an 8mm drive, it might bite in enough to make it seem as if it’s a 7.5mm drive…
An 8mm drive will feed through filament faster than a 7.5mm drive - so - with PLA more material will be extruded with the same E-steps…
Taking this ‘variable drive diameter’ into account, different colors / brands of PLA might behave differently to one another, simply because one is a bit softer and the drive diameter is reduced slightly (and thus the steps per mm needs to be raised).
I just check e-steps with one material, and use ‘flow tweak’ in my slicer for fine-tuning other materials…
REALLY e-steps can only be correct for a certain material… so you would either need to change the e-steps OR change the flow rate for softer / harder materials (so when you chop and change them all the time like me, you don’t stuff around with firmware, you stuff around in the slicer cos it’s easier ;-))
@Jarred_Baines a lot of what is in that link may have been true in early designs but as calibration and slicers have improved it has become less relevant. While it is true that using a hobbed bolt necessitates calibrating e-steps to ensure you are getting the proper amount of filament that is called for, once that calibration is done you really don’t need to be messing with it and reflashing your firmware for each filament change. The reason measurement is needed in the first place is because the accuracy of the hob is different from bolt to bolt. The differences in how PLA, ABS, nylon, etc. are gripped are negligible. It is better to set the proper amount of e-steps, using a long enough extrusion to mitigate error and then use proper measurement of the filament for input into the slicer. You can tell that there is something set wrong somewhere by looking at the photo. Some slicers can seem to “smooth over” errors but that only masks the underlying problem. If the printer’s extruder is calibrated within a few steps then there is a good chance that something in the slicer is set wrong. The only other option is that the filament is outside of spec.