there seems to be a couple of tasks to do now. From your photos, a software (or rather settings) problem is more likely at the moment.
Firstly, the cube is not very large. The printed material (I presume it’s PLA? ABS would be a bit different.) has to be cooled down before the next layer can be printed on top. For larger models, air convection is enough to achieve this, but for small (i.e. fast) layers, the fan has to be used to cool the print. However, the detailed settings for this (fan speed, feedrate reduction, time limits etc.) need to be adjusted depending on your printer, filament, model, … You’ll need to test a bit to get a feel for this.
Secondly, if you have not done this already, calibrate your printer. Most importantly, the amount of filament (i.e. the extrusion multiplier) has to be fine-tuned to get good surface quality. There are already several posts about this here in the forum, as well as on the RepRap wiki, Thingiverse, and many more sites. If you have questions about this, feel free to ask.
Apart from that, there are a couple of settings to play with. Like with the cooling, the main goal should be to learn about the various ways how the (Slic3r) parameters affect printing. E.g. for those thin strings, the retract settings could be a point to start with. Printing speed, acceleration, layer height and extrusion widths are important for other properties, like surface quality. When testing, keep a “fallback config”, e.g. from the standard.ini. It’s easy to alter so many parameters, that in case of low print quality it can be faster to start over than trying to revert the last changes to the config.
When you say it, it surely looks like too much filament is extruded too fast.
Although I’m compleletly new to 3D-printing I also think it is a matter of calibration and tuning. I just wanted to make sure that there is no obvious mechanical problem. And yes, I use PLA, Velleman-brand.
I will make a larger test object and take it from there, following your suggestions, starting with cooling and the extrusion multiplier.
BTW, 3D-printing is in fact quite like machine embroidery (2 1/2-D?), also great fun, where also a myriad of settings and the properties of the material affect the outcome.
Hi. I had the same problem. In slicer… configure… filament settings lower the extrusion multiplier. I’m currently using 0.91 with the Velleman black PLA.
It is all in the settings indeed, on this forum you only see the problems but yesterday we printed something which I am so pleased with that I will try to make some detailed pictures and post them here. We did use the slicer from Cura, since I have read that this is better than slic3r on this forum, and I haven’t tested it again with slic3r to see if it is realy true. But, as said, I am very pleased with the result and now we can go on with even more fine tuning perhaps and finally designing and printing parts.
I have tried various ways to improve the print result, but it seems that the main problem was indeed mechanical due to rough movement on the x-axis, causing the filament to come out in spots, rather than a string.
I tightned the x-belt a bit, oiled the rods (again) and the result is, if not pefect, much better. See the picture below.
I also noticed that the bed was very easily stopped by hand under manual control, but only when moving left on the x-axis. I have never lost a step when printing though, but I raised the voltage for the x-motor to 0.5V to be on the safe side.
Next I will focus on the z-axis. Watching the printing process, it is easy to see that the layers come out differently in consistent cycles of three or four layers, causing the horizontal bands.
I have a barely visible wobble on the Z-axis, so i have left that alone for the time beeing.
Following kuraasu’s excellent instructions in Slic3r settings, and some other fixes I have printed the 10x10 mm boxes below in PLA.
The walls are one line thick and the layer height is 0.4 mm, quite an improvement i think.
One flaw is that I get a small blob at the beginning of each layer that can be seen on the boxes in the top picture that are printed with Randomize starting points on and off respectively.
You could try slicing with cura and load the gcode into repetier, it worked very well for me. All blobs disappeared just by using cura instead of slic3r. If you try, please let us know how it worked for you.
( btw 0.4 mm is quite large I think, I used 0.1).
I also get the best result from cura(maybe because i don’t know enough about repetier). But why import that in to repetier.
I print direct from cura and it works great, there are also a few useful plugins available.
Slic3r’s specialty. It’s a bit reduced with the E3D hotend, since retract works better, but not completely. Have to look into this a bit more myself, maybe including more calibration like suggested here.
Indeed. Nozzle diameter * 0.8 should be the max layer height, so this is the limit for the stock 0.5 mm nozzle. Lower layer heights should give even better surface quality, if the extrusion width is set large enough.
[quote=“2stimpy”]I also get the best result from cura(maybe because i don’t know enough about repetier). But why import that in to repetier.
I print direct from cura and it works great, there are also a few useful plugins available.[/quote]
I used cura but I missed some manual controls like extrude for 10 or 20mm to prime before printing and turning the fan off manually. And I had a lot of error reports, does cura send some not supported commands? I am looking for the plugins right now.
About the Z axis wobble, Im not sure I understand how with the K8200 that would be a big problem. With some other printers where the extruder moves rather than the bed, say a Prusa Mendel, a bent Z axis would almost certainly cause the extruder to move in X and/or Y direction, but with our printer, the Z axis slides over 3 rods and there is virtually no play in them.
That leaves a variation in the Z axis. If the Z motor can move (like with the original incredibly rubbish mount), I can see the problem. Every micrometer the motor moves down is one micrometer that the extruder does not, while it should. A sturdy printed Z motor mount is therefore a very good idea (one of the first things I printed), but a slightly bent shaft, I dont see how that would make as big an impact as many people seem to think.
Thank you all for all your input in such a short time, I really appreciate it!
I am new to 3D Printing, less than one month “experience” including building the thing, but I know from machine embroidery and vinyl cutting that it is essential to first establish a stable and predicible setup and then from there start optimizing and fine tuning. I’m amost there…
I was quite happy about the prints in the pictures above after 30+ test prints but your replies tell me that I can do better.
[quote=“Cartman”][quote=“2stimpy”]I also get the best result from cura(maybe because i don’t know enough about repetier). But why import that in to repetier.
I print direct from cura and it works great, there are also a few useful plugins available.[/quote]
I used cura but I missed some manual controls like extrude for 10 or 20mm to prime before printing and turning the fan off manually. And I had a lot of error reports, does cura send some not supported commands? I am looking for the plugins right now.[/quote]
I don’t prime, i set the rim count to 5. That way when i start the actual print it is primed perfectly with just the right pressure.
There is a nice plugin that lets you set things at a certain height.
I also get lots of errors in the log, and sometimes the printer pauzes for a moment leaving a blob in the print.
Anders s, keep working on it indeed, and you will see improvement.
Yesterday was a test day for me, burnt up the whole day.
it all depends on what “slightly” means. If the rod is nearly straight, the effect will be minimal. But with some curvature, you’ll move the M8 nut supporting the Z axis with each turn of the rod. Thus, the contact surface between the nut and its enclosure will move, too, e.g. if the nut is tilted. Again, as you said, a few microns at the rod position are enough to affect the extruder height by a few hundredths of a mm, causing wobble.
Hi,
There is my solution to reduce Z-axis wobbling:
I use larger screws and turn original part upside down. In result there is enough space to compensate z-axis screw horizontal movements, but it perfectly holds vertical position.
[quote=“drummy”]Hi,
There is my solution to reduce Z-axis wobbling:
I use larger screws and turn original part upside down. In result there is enough space to compensate z-axis screw horizontal movements, but it perfectly holds vertical position.