3dmaker4u
-
Posts
102 -
Joined
-
Last visited
Content Type
Forums
Events
3D Prints
Posts posted by 3dmaker4u
-
-
George, seems that @VISU-AL has a SF UM2, so he knows what is talking about.
And it is something related to the topic, which proved to be actually not an issue with top layer only, but also affecting other layers.
And the story is that solid layers print fine, but not with 100% infill, for a reason that nobody actually found.
And this appear to affect new machines.
Forgetting about any axes misalignment, I begin to suspect a kind of limitation of the extruder speed. I don't see signs of underextrusion in the sense that is normally used. What I can see is that while the user is expecting that the settings will produce, at normal flow, a particular extrusion width (required for 100% infill), the result is less flow than required.
So, at 0.1 mm it works, but at 0.2 mm (and com'on is only 4 cmm/s, not a reason for underextrusion at 240 C
there is not enough filament extruded, but no skipping or any other obvious underextrusion reasons.Looks like the E speed and/or E steps/mm does not correlate to the actual printing speed.
-
Well, since you are there anyway to change the filament, you can manually command, but definitely a pause (preceded by a head move) will definitely be better and all the codes (though sliced separately) could be "cleaned" (remove unwanted/unnecessary instructions) and assembled together.
Very interesting idea!
-
If you are in the chapter "how can I use thingies?", what about using some... pins
Cut the two ends so that you get a pretty hard transverse. Then you can heat end and then just pin it in the filament. It will automatically glue. The other end could be trickier. Perhaps a soldering iron could be used to heat it, but you also have to provide a thermal break, e.g. a clamp attached in the middle.
Alternatively, just lay down the two parallel filament wires and press a heated pin on both.
Just a couple of thoughts
-
It's too hot for the actual speed you're printing the top. Options:
- decrease temperature in general (this is useful for the other problematic areas)
- decrease minimum speed in cooling control
-
Do this in your slicer, by adding the corresponding "end Gcode". Octoprint can only help in let you manually control the HE and HB temperatures (preheating, if you wanna say so).
-
I would suggest you should think splitting the model in two parts, and print them separately then glue together. You cad do either face-back or left-right/ If doing so, use also a generous brim to prevent warping, and cut it after gluing.
For the holes on top of the head, increase top solid layers and/or general infill. For 15% infill, use at least 6 top solid layers for 0.15 layer height, or 8-10 for 0.1 mm layer height.
-
To the first point, I would say you're printing either too hot or too cold.
-
90C bed temperature is by far too low for ABS to keep sticking on the bed. Start with 110C. You can eventually increase to 115C or even more. This will depend on the accuracy of temperature measurement, the gradient of the temperature on the surface of the bed and the air flow around.
Use a brim of at least 3-4 mm. You should increase it pretty much (8-10 mm) if your part has acute angles or tall vertical edges. In general, the larger the part, the wider the brim. However, it is important that the brim sticks well by itself and no air bubbles are trapped underside. For this level your bed so that is not too high, but not too low either.
Try to avoid any air flow and use the cooling fans only if your part has serious overhangs in your part.
Glass is not very friendly in helping you with ABS. If you find that none of the above helps you, cover the glass with Kapton tape. This will help a lot and will allow to use the fan even after a few layers. Still keep the bed heated at 110C.
-
Well, is not only with CT. It's scanning in general.
The scanner itself is helpless, except if it is accompanied by a good post-processing application.
This has to intelligently smooth the model, clean it up and do some other stuff for the model to become useful for other application.
This is however application dependent, since different application will have different requirements.
I don't know the requirements for the output generated by the software used with the CT, but it definitely not appropriate for direct 3D printing.
http://umforum.ultimaker.com/index.php?/gallery/album/1439-medicalmodeller-bone-printing-issue/ are some images underlying what I wrote in my previous post.
-
Ok, just had a look at the models.
It appears they are the result of a kind of scan, perhaps X-ray or something alike.
The bad news is that your model contains lots of detached parts. This can be easily viewed in MeshMixer (was suggested before as an alternative to things you cannot do with netfabb for no money) by doing Analysis > Inspector (lots of magenta spheres). That's something that has to be cleaned up, though this is not your problem
It is also that the interior of the bones has many crazy details that for sure are not fragments of the broken bone, but just the result of scanning. It's most probably not the scanning device itself, except for the limitation of its resolution, but of the post-processing software that does not an "automatic cleaning" of any sort.
And here starts the trouble. All these nitty-gritty produce lots of retractions. Correlated with some lack of fine tuning of the printer and of the slicing parameters, I believe that your hotend is clogging. I don't know exactly how UM handles clogging once it happens, but I expect that, at least for your case, the filament will be grinded, and hence no skipping, but also no feeding.
I would recommend:
- clean the model of detached parts (Autorepair all)
- reduce complexity (as pointed out by Didier above) recursively, as many times as you cannot seize a significant distorsion ( I think you can easily get it to 200-400k triangles)
- smooth the interior of the bone as much as possible (can be done with sculpting tools of MeshMixer); this should be manual to avoid deforming things of relevance
- slice it again
- inspect the gcode layer by layer and see how simplified/complicated looks (you should be able to compare with the initial gcode)
This is all on the model side.
There might be issues though with the printer and printing parameters tuning (for instance, I don't get why do you use 0 infill).
-
Yeap, I agree, though I was referring to the impact on print
However, about uneven heating, are there real chances with either a PCB heater or a silicone pad and a sheet of aluminum to have such bad variance of temperature to cause breaking the glass?
-
Happy to dropbox the stl if someone wants to play and see where I am going wrong.........
Please, do it
-
The controller mount should be decomposed in 2 brackets and 1 support that would attach to the brackets through a push-slide configuration. This way you can print each part in a "natural" position and have no overhang/support issue, not to mention the mechanical resistance problem that the original design expose.
The re-design could be easily done in 123D Design, for example. Just import the STL from Thingiverse as a model and start designing the 3 individual parts.
-
I don't think thermal expansion counts at all. Printing is achieved at rather constant bed temperature. Well, first layer may be at higher, but normally just 10 degrees, and contraction from say 65C to 55C will be something like 0.01 mm per 100 mm.
I had a similar concern for a quite big ceramic plate (ca. 500x500 mm) on an X400 from GermanRepRap. But when I measured the variance between 20C and 110C, it was practically zero.
"Adhesion" may definitely differ. I think there are two kinds of "adhesion", when you do not use any kind of "glue".
1. By normal air pressure - this is obtained if the surface is as of... glass (glass, kapton, PET). On such a surface, if the plastic is laid down correctly, leaving (near to) no air underside, the atmospheric air pressure will keep the part on the bed. This is why, in such a case, you can detach the part by hitting it laterally, making it to slide on the surface and get some air that will actually make it... fly
2. Through porosity - these apply to such surfaces, as in the case of blue tape, BuildTak. There are also some "profi" machines that uses intentionally porous bed materials in different forms.
One interesting point is about the bed temp. Basically, the rule says that first layer should be above glass transition temp, say 5C, while the rest should be just slightly below, also 5C just for simplicity
The only issue is that not always the actual value is known or is exactly the one stated by the manufacturer, but is a good starting point. To calibrate, one should try to find the temperature where the plastic become hard from soft and/or the other way around. -
It's peculiar anyway.
If it should have been bent, the pattern would have normally appeared in the center only. As it looks is rather that the rods are... oval!? Eccentric pulleys would produce the same effect, since they will tend to bend the rods that are fixed at the ends.
And if the rods are oval then the pattern will be repeat on each layer, but differently I think, due to the 90 degrees change in the orientation of the infill and to the very little difference between the starting and ending point of the perimeters on each layer (at least for the very simple square shape in the test). As a result, the different thickness/width is not appearing only in the first layer, but in all the layers. If this is the case, than printing a "flat cube" with 100% infill will perhaps expose this peculiar effect on all layers.
Pure induction and I might be wrong
-
How many teeth has the GT2 pulleys? 20?
Seems that, for a reason or another, at each turn of the smooth rods there is a slight lack of parallelism between the pairs, hence the carriage is moving up and down on a 20x2 mm pattern on both X and Y.
-
The left one looks very much like ribbing produced by layer height vs. leadscrew pitch.
-
So, in most pictures the scar is "negative" (some missing plastic), but I saw at least one that is "positive" (some more plastic than necessary).
The factors that influence the scar are:
- retraction speed/acceleration: at the end of the outer line of the shell (exterior or interior) retraction takes place before moving to infill; if the retraction is too fast, a bit of the already extruded plastic will be sucked back, leaving kind of a hole - the negative scar; if the retraction is too slow, the nozzle could ooze, leaving kind of a bump - the positive scar
- printing speed/acceleration: if it is too fast, too much plastic will come out during deceleration, due to the pressure inside the nozzle that could not be reduced at the same pace, resulting in positive scar
- travel speed: if retraction is too short/slow, fast travel may pull a bit of extruded plastic -> negative scar, while slow travel will let more plastic to ooze -> positive scar
- temperature/cooling: this will influence viscosity of the extruded plastic hence will boost or dim the above effects.
And we can list more factors that contribute. So, there is no simple setting solving the issue. However, thinking og negative scar, you may want to:
- reduce retraction speed and/or acceleration
- increase (!) printing speed and/or acceleration
To complement/compensate the produced effect, you may also try to adjust travel speed and/or temperature.
But first, try adjusting retraction. Since you are printing one object at a time (I may assume that from the printing time you mentioned), there is little need for 60 mm/s retraction speed at the quite low printing/travel speeds.
-
Ooops! That should read 0.12-0.15 mm
-
That's an issue caused by overhanging.
The point is that, if on a particular layer, the perimeter starts to curl in a region, it will evolve vertically, and perhaps extend horizontally as the layers build up. There is definitely more heat in the affected area. Could be that retraction takes place nearby, before jumping to the next layer.
As with any overhang, the factors affecting the result are temperatures, cooling, speed and layer height. You already have good values for the parameters, but seems that they did not suffice for the filament you are using and/or the part you are printing.
Also note that "screening" the bed may also have side effects. Since the heat has to escape somehow, it will find the "least resistance" path. This could be along the edge of the "screen" that is on the printed part side. Which is something you don't want.
I would try a slight increase in layer height (1.2-1.5 mm).
-
It looks to me a combination of:
- oozing filament (you know that color can make the difference, even for the same manufacturer)
- too high temperature (especially true for the above, where a few degrees may count)
- too much fan
well, I thing this is how the fur is spreading everywhereThe images remembered me about cotton candy, produced from sugar melted in a heated drum, blown by a fan and collected as "fur" on a... stick
-
I think there are two perspectives we have to distinguish here.
One is related to "design for print", understanding you keep considering both the printer and the slicer, when you create a part. This is taking info from the printing technology and from the slicer manufacturer and take them into account as "design rules".
On this side, a 1.6 mm wall, printed with a 0.4 printer, using the Gcode produced by a slicer that uses nozzle diameter as the width of the final printed extrusion should be guaranteed to produce 4 perimeters with no infill. There could be another parameter for the extrusion width, as in case of Slic3r which almost ignores nozzle diameter in calculation (well, not completely). However, it seems that there are numerous situations that developers should consider in the routing module, and trying to accommodate all sometimes produces peculiar effects.
So, we are coming to the other perspective. If strange behavior is observed, it has to be sent to developers and they have to come with arguments and/or to review the routines.
If we leave aside for a while the "square spiral", a cylinder with a 1.6 mm wall, sliced in a way (no matter which) that should "guarantee" a 0.4 mm per printed track, must have 4 "synchronized" shells (perimeters). This kind of synchronization might be tricky, but is obviously required if printing is to be optimized (i.e. the slicer to be considered of... high performance).
To me it appears that lack of "synchronization" between segments, which should be otherwise parallel and of "proportional" length, so that they will produce "concentric" shells, is what produces the "need" for infilling. And this probably apply to any shape that has round parts (I doubt that the problem appear on a "pure" cube).
To resume this otherwise short post
I think that if no other parameter in Cura configuration is to be used in "controlling" the ratio between shell thickness and extrusion width, then the issue mentioned in the first post has to be addressed by Cura developers. And if such a parameter exists (e.g. any applicable overlap factor), it's influence should be documented somewhere.Finaly
I don't think overlaping apply for perimeters (shells). By the very nature of FFF technology, adjacent extrusions "expand" when printed so that they will "fuse" together, while infills may need some additional overlapping to compensate for the more reduced contact area with the shell. -
If you want something more "visual", then in MeshMixer:
- import your mesh
- press W to activate wireframe view, so that you can see the faces
- select Sculpt
- Select Brushes > Reduce
- "Paint" on the model
- Play with Properties, mainly Size and Strength to adjust the effect
- Press Ctrl+Z to undo the change you don't like
-
It appears to my eyes that different colors provide different levels of contrast when you observe a 3D printed object. If this is true, the question is, which colors are the best?
This is not about observing the artifacts. By contrary, this assumes you obtained a nice smooth surface, perhaps as mate as possible, and you want to analyse details of the object.
I tried to search, but cannot find relevant keywords to avoid bulk results about contrast between colors, so I said to myself that someone may know.
Obviously is a matter of perception, it is subjective, and may vary between observers, but could be that statistically some colors perform better.
To me, for instance, mid-tone colors provide better contrast than lighter (e.g. white, yellow) or darker (e.g. black).
there is not enough filament extruded, but no skipping or any other obvious underextrusion reasons.
I don't think overlaping apply for perimeters (shells). By the very nature of FFF technology, adjacent extrusions "expand" when printed so that they will "fuse" together, while infills may need some additional overlapping to compensate for the more reduced contact area with the shell.
Top Layers not touching - UM2
in UltiMaker 3D printers
Posted
But what if the firmware received some wrong configuration and the extruder does not want to feed more than a value. Which is not necessary a firmware parameter.
This behavior was reported on new printers only and in all cases it's about the same behavior: reduced infill when it should be 100%, most visible on top layers, but not present on first layer.
Gcode was checked and if VISU-AL codes will confirm, gcode reasons are excluded.
Then only the firmware could make the difference.
Don't know how production flow is, but it does not have to be for all printers.