Jump to content
Ultimaker Community of 3D Printing Experts


  • Content Count

  • Joined

  • Last visited

  • Days Won


Everything posted by geert_2

  1. Hello again, I have tried smaller gaps than 0.2mm too, but that did not work well with my filament (Ultimaker PLA and colorFabb PLA/PHA) and my models. If the gap is too narrow, the bottom layer of the model sticks too hard to the support. When trying to remove it, this severely damaged the bottom layer of my models. With a gap of 0.2mm or 0.3mm, and the ribs rotated 90° as shown (very important), this gave the best balance between an acceptable bottom layer, and ease of removal of the support. At least for my relatively small models. For bigger models, I might try a bigger gap of 0.4
  2. I have seen brown, grey and black debris, but no green yet. These were all from burnt material inside the nozzle. If you do an atomic pull (also named "cold pull"), is any dirt green too? It could be that some of the white additives decompose, or react with the decomposing PLA and turn green when too hot? What you could try is keep an end of filament in a flame until it melts and starts burning, and see what color that gets?
  3. PLA does get dull and brittle. Out of curiosity, I tried a filter in the waste water syphon of my lab. It took a year or so, and it still functioned, but you could clearly see the dullness and color change (more pale, white-ish). So there is some degradation going on. I haven't tried any other materials, but I would guess that PET or its derivatives are worth a try. They are used for lots of drink bottles anyway. If you select a water-clear filament, you can easier see what is going on inside the material (e.g. color changes).
  4. Just winding it up again might work, while gently guiding the filament by hand. If it is not tangled too much, it might just snap back onto the spool with a little help. At least it does so for my PLA. Otherwise remove the spool from the machine, walk it down your room, hall, garden and street until the filament is all straightened out, and from there wind it up again. Or cut off the unwound piece from the spool and use it up like it is.
  5. If the problem is only on small parts, the cause might be not enough cooling time for each layer, and slight overextrusion due to the print head having to slow down and change direction very often. A solution you could try: print slow and cool (e.g. for PLA: 180°C to 190°C, and 20mm/s). And print a dummy block or tower next to your models. So the printer has to spend some time on that dummy block, and the real models get enough time to cool down. You could also try to set the flow rate to 95%. In my experience this helps for small parts.
  6. What you could also do, is keep the spool of filament locked in a sealed bag with disseccant (if I spell that correctly). And only take it out to unwind and cut off the required amount of filament. Then immediately put the spool sealed away again. And only feed the cut off end into the printer. The disadvantage is that you need to calculate some spare in, and the piece of filament in the bowden tube is always lost. You you have a lot of waste pieces of filament. But this might cost you less than destroyed models, especially in a commercial environment. I am not a fan of wasting material, but i
  7. In my experience, there are two main causes: 1) For good bed adhesion, the first layer needs to be squished hard agains the glass plate. This tends to spread that layer out a bit, thus widening it. Solution: calibrate the build plate for a higher gap between nozzle and first layer, so that the first layer isn't squished so hard. 2) The heated build plate makes the bottom layers melt and sag a little bit, since the optimal build plate temp is around the glass transition temp of the material (= the temp at which it begins to get weak). Solution: use a lower build plate temp. Both solutions gi
  8. In addition to the tips above from cloakfiend and GR5, if the models are small, you could also try printing them with 100% infill. For better bed adhesion, you could also try the "salt method": gently wipe the glass plate with a tissue moistened with salt water, prior to printing. In my experience this greatly improves bonding of PLA to the bare glass. (It does not work for ABS, and I don't know yet for other materials.) See the full PDF-manual on this salt method at:https://www.uantwerpen.be/nl/personeel/geert-keteleer/manuals/
  9. Forgot to say: Another method that sometimes works, depending on the model is this: - at the bottom touching the glass plate: a solid layer of 0.5mm thick. - then a gap of 0.5mm - solid layer of 0.4mm - gap of 0.5mm - solid layer of 0.4mm - gap of 0.5mm - and so on... - solid layer of 0.4mm - ribs of 0.5mm wide, 90° rotated to the direction of the fill-in of the bottom layer of the model, as shown in the pic above. - gap of 0.2mm to 0.3mm - rest of the model above the support So the support structure consists of separate layers above each other, all "floating" in the air. Due to t
  10. If you design the models yourself, then also design the support structures. Think about these points: - Design hooks, holes or other features into the support, so that you can pull them out with pliers, screw drivers, or other tools. - Allow enough room for these tools. - Allow enough room for a sharp knife or scalpel (surgical knife) to move in-between the supports and the model, so you can cut the supports loose from the model. - In difficult to access areas, extend the support to outside of the model for easier access. So you can easily grip it from the outside with a plier. - Between
  11. That Onshape change, is that only for new customers? I already had an account and I see no changes in behaviour? Private documents still seem to be private, and I can still make 9 private documents. Yes, nine private documents, not ten like before. I don't know why, but Onshape somehow claims that I have one document when I have none, so I can only add 9. Anyway, I rarely use Onshape, since I can not find my way in the user-interface: functions like opening and saving files are splattered all over the page at random, there is no consistency. At least, I can not find the logic in it.
  12. The only thing for which Sketch-Up has some limited use in design for 3D-printing, is when your 3D-editing package does not support some features and Sketch-Up does. For example text in DesignSpark Mechanical is not officially supported (although it can be done by adding and editing dimensions, and then projecting these onto a model). In such a case you could set the text in Sketch-Up (only the vectors, no 3D), and import that in DesingSpark Mechanical. But a lot of characters will need manual fixing, since sometimes the vectors do not touch each other, leaving a gap, and then you can not ext
  13. Hello Cloakfiend: a way to release complex shapes from a silicon mould might be to use a thinner mould. Like they do when casting human bodies. A thin layer of silicone can be removed like a blanket. But then you need to design and 3D-print a tight fitting hard shell too, that can be opened, and with register marks, so you can fit it together exactly. Otherwise the silicon mould does not keep its shape. This also saves you a lot of money on silicone, because you need far less. Google for Youtube videos on mould making and casting. Another thing that helps, is to spray the model with silicon
  14. As Nicolinux says: even within one brand and product range, some filaments have totally different stringing characteristics. For example in PLA/PHA from colorFabb: the "olympic gold" and "natural" colors do string more than standard white I had. Probably due to the amount and sort of fillers used? Ultimaker's "Pearl" (nice color by the way) also strings a lot. But they do stick very well to the build plate. ICE PLA strings far less, also prints nice, but does not stick so good to the build plate. Also, the sticky and "stringy" materials are more difficult to grind: the molecular structure ap
  15. You could also have a look at shops for custom car parts, spoilers, wheels, accessories,... They usually have a lot of special metalic paints for plastics, and they should know how to get it to stick to plastic (how to grind, how to chemically activate the surface, apply primer, and so on).
  16. I don't know anything about Blender either, but is there no way to set physical dimensions? Or, if it would be a sort of "freehand sculpture", without any reference to dimensions, maybe he could create a little cube next to it, and give that cube exact physical dimensions? Since people can design planes exactly to scale in Blender, something like this should be possible.
  17. What I would try in such a case: - Thorougly clean the outside of the nozzle prior to starting the print, and then wipe it with silicone oil or ptfe oil, to reduce the build-up of molten filament during printing: this gives less "hairs" and less strings. Make very sure to not spoil any ptfe oil on the glass plate, or you will have no bonding at all. - Gently wipe the glass plate with *salt water* to improve bonding and prevent warping (it seems you have some corners lifting?). This works for PLA only, as far as I know. - Print speed: 20 to 25mm/s. - Nozzle temp: 190°C (start the first laye
  18. Even if you don't need it for real work anymore, why not keep the UM2, and use it to experiment with all sorts of things? Try different filaments, different bonding methods, try upgrades, whatever. Or use it as a demo printer for friends, collegues, kids,... to make toys or so?
  19. For relatively simple technical parts based on geometric shapes, and where you often need to adjust dimensions, DesignSpark Mechanical (freeware) might be a good choice. Easy to learn, lots of good training videos, and you can also use SpaceClaim videos, since DSM is a limited subset of this. For complex organic shapes, I think Blender (also freeware) might be a good choice, but it has a *huge* learning curve. You might also have a look into Form-Z from Autodessys: not free, but they have quite a lot of functions for organic shapes. I once tried an old demo a couple of years ago (was 1 month
  20. I had a couple of typical 3D-models printed on an expensive Objet printer, which works similar to inkjet printer technology: a set of jets spraying liquid plastic, and an ultra violet lamp follows and cures it. Resolution is a bit better than on my UM2, especially for text on the top layer. But it is not exceptionally. And text on the sides is not readable at all. You still can't compare it to injection moulding quality. After each use, the printer requires at least 30 min of cleaning, otherwise the jets may block (1000 euro per jet x 8 jets). Resin costed about 1500 euro per liter. Support
  21. Can't Blender or FreeCad open and repair them? They are both freeware, so that might be worth a try?
  22. The steel wire version of the FilCatch won't melt or break. I have been using this now for a couple of weeks, and it works well. So this was a very simple but great idea.
  23. I tried this today, but it did not work well. What I feared did happen indeed: molten filament creeps up between the nozzle and silicone cover, and it breaks the silicone at the top (where the nozzle meets the aluminum fan mounts) and pours out there in a big blob. This is probably due to the silicone cover being flush with the nozzle, so it sort of pushes a bit on the surface while printing. When there is a little bit of overextrusion, as on the first layer or on short strokes with frequent stops, the melt choses the way of the least resistance, which seems to be in-between the nozzle and si
  24. Hmm, the first annealing test in the Ultimaker was not totally convincing. In the photo above, the flat object is freshly printed, the warped one is annealed. Both are the same PLA. While it was sitting in the "oven" (=under a cover on the ultimaker bed), I couldn't resist playing with the temperature, and I increased it to 80°C, so it came loose from the build plate and reverted to a more relaxed state. I should have left it at 55°C, I guess... Anyway, this shows how high the built-in stresses can be, even in "low shrink" PLA. And what you might get when you leave such a model in your car
  25. I have similar font weirdnesses as Daid describes, although not identical. It looks a bit like the font is drunk. Also the fonts in the title bar (Explore Products Stories etc...) look pixelated. OS: Windows 7 SP1 Browser: Pale Moon (most recent version). Pale Moon is a split off of Firefox which still uses the old-style layout with menu bars and status bars.
  • Create New...