Jump to content

geert_2

Ambassador
  • Posts

    2,084
  • Joined

  • Last visited

  • Days Won

    34

Everything posted by geert_2

  1. I think effects like this might also be caused by fluctuating temperature, because that changes viscosity and flow-rate (=less back-pressure when more liquid). And maybe also by a worn-out white teflon coupler? But I don't know a good means to reliably measure the temperature from the outside? A thin and heat-resistant sensor that you could move into the nozzle from above might be best? (Of course after removing filament, removing bowden tube, and cleaning the nozzle.) And then let the printer run, "printing" this same object. Thus it goes through the same moves, same temperatures, same heated bed, but only without filament?
  2. I would suggest: cut out a small part of this model with some text for testing purposes only, so you don't waste time and material doing the whole thing. Try printing at 25mm/s, cool (but without underextrusion), and in thin layers (0.08mm...0.12mm). See what that gives? 3D-printing is always searching for a balance between quality and speed, but often they don't go together. So you will have to chose your priorities... In my experience (with UM2) such text on the sides always comes out poor, due to ringing effects and widening in corners when slowing down. And you never get sharp corners anyway. Text on top of flat surfaces comes out *much* better. So if you could move the text to the flat horizontal plate halfway (if that doesn't get covered later on), this might be an option? Photos: Text on top of model: these characters are ca. 7mm high. Orange PLA. Left: smoothed with acetone. Center: heat-treated with hot air gun in attempt to smooth (this damaged the model, so don't do this). Right: untreated, as printed. Printed at defaults for PLA: 50mm/s, 210°C, flow 100%. Printing speed and layer thickness tests: Top row printed at 50mm/s, bottom row at 10mm/s. Nozzle 0.4mm. Material is transparent PET. Model is 20mm x 10mm x 10mm. Watermark text (hollows) is 3.5mm caps height. Layer-thickness from left to right: 0.4mm, 0.3mm, 0.2mm, 0.1mm, 0.06mm. So the only way to get good quality, is by printing slow and in thin layers. And then reduce temperature to prevent overheating, leaking, or decomposing of the material in the nozzle.
  3. I like the girl model. Have you ever tried painting such models in a gentle sort of "marble" technique? So it looks like stone or marble? I think that might work for this girl, and it might conceal layer lines and defects. I once knew an artiste who turned lightweight wooden crates into "heavyweight marble" bases for displaying sculptures. These bases were hard to distinguish from real marble. But I don't know how she did it, I only saw the results, not while applying it. Might be worth trying that? I also like the "old" look of the partially painted and sanded model. Maybe that would also be an option: paint multiple layers in slightly different shades of grey, and then sand off a couple of them? That too would help conceal printing-lines, in the way military camouflage hides the shape of vehicles or aircraft.
  4. I think it will be hard to find something that has the flexibility, chemical inertness, long life, temperature resistance, stickyness and grip of silicone. What you could do, is print a mould in PLA (or whatever), and pour your own silicone sleeves in it. Then you still need silicone of course, but at least you have the design flexibility you want, that off-the-shelf part can not give.
  5. I would suggest you embed the pictures in your message, instead of using external links. Then you are more likely to get answers. While writing your text, place the cursor in the desired spot, and drag and drop the picture from Windows Explorer into your message. I guess this should also work on a Mac or Linux. Due to the enormously increased phishing and hacking since the corona-crisis, people are less and less willing to click on links from unknown sources.
  6. For those who print in PLA: do not leave these masks in your car in the sun, or behind any window. Even not on mild spring days. Because it *will* deform due to the warmth, especially when under mechanical load: PLA starts getting soft from ca. 50...55°C, and that is reached in a car as soon as the sun comes out. Don't ask how I know. :-) For the same reason: Do not clean the masks with hot water or steam jets, and do not dry in hot air. Comfortable handwarm water is the maximum. Also explain this to the people who are going to use those masks, and leave a written note in the package. Most likely those people are not aware of this, and they are going to expect the plastic to be similar to Lego bricks (ABS) or or plastic bottles (PET), the plastics they are used to. Which it is not. Also make them aware that these masks do not protect against any bacteria or viruses. They only protect from direct spats (blood, spit, pus, acids,...) or from objects flying directly towards you (when drilling, sawing,...). Even though this seems totally obvious for us technicians and engineers, it doesn't occur to a lot of non-technical people. Every day I see people wearing mouth masks that are no better in filtering viruses than chicken wire or mosquito nets. Or sometimes they have a suitable mask (FFP2 or FFP3 without exhaust valve in Europe, which I think is similar to N95 or N100 in the USA?) but wear it only covering their mouth, and then they breath through their nose of course... While endlessly queueing to enter a shop yesterday, like in the heydays of communist DDR during the Cold War, I saw tens of people wearing masks, but not one wearing a correct mask in a correct way. Not even one. Seems like the virus and hysteria in the first place destroys their brains and common sense... However, for me, it seems it infected my tolerance for stupidity. :-)
  7. Hmm, synthetic plastic food and colorants. Can't be bad. :-)
  8. Another thing that might help, is when you keep sitting next to the printer. Then when it is reaching the top of the models, gently and gradually lower the flow-rate of the extrusion, via the Tune-menu on the printer (name might differ on your printer, depending on the model). Reduce flow-rate very gradually to 95% or 90%, try what works for you. Also, manually reduce temperature as it reaches the top; and here too: try what works for you. The shape will still not be good, but it will overextrude less. My test demos were printed with about standard settings, thus way too fast and too hot for these tiny cones. But they do show the effect very well. :-)
  9. These "insect antennas" come from the nozzle leaking while traveling through air. Upon arriving at the next wall, it deposits that leak as a drop on the side of the wall. Upon the next layer, the drop is deposited on top of the existing drop. And so on, causing a nice insect antenna. If you watch closely while printing, you can see the antennas growing with each nozzle pass. So you need to tune your settings for less leaking while traveling (print slow, cool, thin layers, enough retraction, travel fast through air), or for not traveling through air at all, if the model allows it. But I have never printed ABS (apart from one test cube), so I can't give more specific tips. Here a few pics: one closup, and the other showing the dimensions of the piece related to other parts and a ruler (in mm and cm). This is printed with a 0.4mm nozzle, in PLA, after I forgot to switch retraction on.
  10. My standard pic to show the effect of a dummy cooling tower (ruler is in mm and cm). In addition to what gr5 said, printing in thin layers, slow and as cool as possible also helps. Also, do not use 100% infill but rather 50% or less. But I could never totally get rid of the effect. At some point the hot nozzle just keeps sitting on top of the tiny model, preventing it from cooling down and solidifying. That is a limitation of this printing method.
  11. A few questions: Are there any soft leather, rubber or cloth straps as interface between the 3D-printed parts and skin, to increase comfort? So that the hard plastic does not continuously rub the skin, causing damage? I have worn safety helmets in the past, and they all had a hard outer shell for protection (obviously), a soft inner plastic frame for good fitting around the head and for shock-absorption, and then a thick, soft leather strap between that soft plastic frame and the skin. Otherwise it was not possible to wear the helmet for a prolonged amount of time, without getting wounded. How is that handled with these visors? (I have also wondered about this for the 3D-printed respirators, by the way: for a good fit without leaks, and for comfort and not getting wounded, they also need a soft liner. But I have never seen it mentioned?) Related: is the inside sanded or chemically smoothed to get rid of the most layer lines, for improved hygiene and comfort? Out of curiosity: what is the purpose of the lower strap? We usually don't see that on commercials masks? The design could also work as well for hobby use like gardening, working with a chain saw, or similar. It does not protect from dust or fumes, but it does protect against direct liquid spats and bits and pieces flying around.
  12. At big oil companies, but I don't think it is a good idea to go buying a vat of 160 liter... 🙂 And there are so many different specs for different purposes. I just use hydraulic oil because I had it for my hydraulic machine (and it is on-spec for that hydraulic machine, not necessarily for any other equipment). If I had to buy fresh oil, only for the printer, I would probably go for something close to the official Ultimaker specs: some light machine oil, like used for lubricating bike chains, fine equipment, tools, and similar. You can find that in car-accessories shops, or do-it-yourself hobby shops (at least here in Europe). No car engine oil (=way too thick), no ultrathin dislodging oil (could be corrosive and does not lubricate well), no real sewing machine oil from a textile shop (this is the one that turned into gum here), no brake fluid (this is no oil at all, but a different kind of high-temp liquid), no grease (too thick), no high-temp oil,... Note that I am not related to the Ultimaker-company, so my words are not the official company policy; it's just my personal experience as customer and printer-owner.
  13. I would suggest you add pictures to your post. Then you are more likely to get good answers. You can drag and drop JPG- or PNG-files directly from Windows Explorer into your post while editing. Set the cursor on the desired place, and then drag the picture into there. The reason is that the amount of phishing and hacking has increased a lot since the corona-crisis, so people are less and less willing to open unknown files from people they don't know very well. Without having seen your design (for that reason), my general experience with snap fit things is that it mostly depends on the material it is printed in. PLA works a couple of times, but then tends to break, while PET will keep working longer, provided that the design is good. But the layers may hinder smooth sliding, depending on the printing orientation. Not a snap lock, but it goes into the same direction. This carabiner in PET (green) keeps working well. In PLA (cream) it tends to crack after some time, and it tends to deform permanently.
  14. Indeed, I haven't seen this one. :-) Seems like I should bring my knowledge about tools up to date. The inserts I have seen on Youtube videos, but not tried myself yet. If you have used them, do they stick well, and does the molten plastic reflow around the ribs well? I have also seen a similar concept, in which the brass insert had to be screwed into the plastic hole, with a sort of self-tapping one-way screw on its outside, so this would get stuck and stay in there. But again, no personal experience.
  15. I tried tapping too with straight taps (I had never seen those spiraling ones, didn't know they existed), but indeed it didn't work well. Even not when tapping manually and reversing very often: 1/4 turn forward, 1/8 turn backwards. Chips would get stuck and the plastic would melt (even at very low speed, lots of cooling time and lubrication). Also, threads were very weak and got worn-out very soon (M3, M4, M5). So for most of my models, I now use standard nylon M4-screws instead, with a caged nylon nut that can not fall out. This goes a lot faster, and it works better: the thread (nylon nut) is much stronger than a 3D-printed and tapped one, and it can be replaced if required. Mostly variations on these concepts: 1. Fully caged nut; the backside of this opening is closed, it is only open on top. If the opening is tight enough, the nut will even be clamped. If slightly wider, it will wiggle, but not fall out (unless you totally remove the screw and hold it upside-down). 2. Another variation on the same concept: 3. Recessed hex opening for keeping the nut. This is a test model to try-out required tolerances for an M4 nut (ca. 7mm diameter between flats). 4. I haven't tried this yet, but someone (I forgot the name) suggested using "metrinch" cages for clamping the nuts. They seem to wear-out less than traditional hex holes, and allow tighter tolerances, since problems with 3D-printing are mostly around corners.
  16. Depending on the firmware-version the "Abort" on the UM2 can be plainly visible during printing, or in the "Tune" menu (I have both). In some UM2 there was an issue with driver chips overheating: then they would shut down for a few moments until cooled down, and then continue. Mostly with the Z-axis (up-down). I had this on one printer a few times in hot weather, when doing long print runs. I reduced maximum current a little bit, and since then it never happened again. I am not sure if this could happen to the X- or Y-axis too? But first try the other solutions, and only consider this as a last resort if the other things wouldn't help. Not only no oil, but wrong oil on the rods could cause issues too: I once had "thin oil" that soon dried into a sort of very sticky gum. So much so that I almost couldn't move the head anymore, it was almost glued to the rods. That caused skipping too. Then I cleaned the rods thoroughly with isopropyl alcohol. And now I use surplus hydraulic oil, as used in hydraulic test benches and machinery. I don't know if this is the best oil, probably not, but at least it lubricates well, has anti-corrosion additives included, and it never dries out. It works for me. I just regularly refresh the oil because of the dust that gets collected on the rods. When lubricating, do not pour oil on the rods, but rather add it to a tissue, and then wipe the rods with the tissue (without touching the belts). You do not want oil dripping all around in the printer: it would destroy bonding to the glass.
  17. I have tried these: - PLA, orange and white: bright colors, printed very nice and smooth, but stuck a little bit less to the glass than Ultimaker or colorFabb PLA when using my "salt method" (=wiping glass with tissue moistened with salt water). Seemed to be slightly less temperature resistant than other PLA. Otherwise no problems. Prints at standard speeds and temps. When burning pieces in a metal spoon on a bunsen-burner, it leaves very little black ash: just a little bit of dry, loose powder that can easily be wiped off. So, the same can be expected in the nozzle. - PET: prints okay, but doesn't like to do overhangs without cooling: when liquid it is more rubbery than PLA, so it does not make a nice string but rather snaps and folds back into a blob onto the nozzle. So it has trouble closing gaps. I don't know if this is brand-related or rather material-related, because I have no comparison. Could be just a general PET-characteristic. I use this PET for printing carabiner hooks (more flexible than PLA, and less creep), for printing watermark logos (transparent), and for stuff in the car (can withstand summers, contrary to PLA). Prints best slower and cooler than standard, I generally use 215...225°C and 25...35mm/s. When burning in a metal spoon, it leaves a thick glossy sort of varnish, that is hard to remove. So, when printing too hot, this might also occur in the nozzle, I guess. Again, not sure if this is a brand-characteristic, or PET-characteristic. Thus print cool. - ABS+: only printed one test-cube with it. Less strong than PET or PLA, way softer also. Warps a lot and smells bad, as you could expect from ABS. I can't compare with other brands, as I never printed any other ABS. But I would use PET or PLA instead. I think the PLA and PET are worth a try. I wouldn't use any ABS, unless really necessary. Models below are in ICE-PET: ,
  18. I think the extra width that you measure might come from ringing- and thickening-effects around corners? When slowing down to take a corner, the nozzle inside pressure does not immediately drop, it lags, so the nozzle extrudes a bit too much compared to the now slower speed. This makes corners thicker. Analog for ringing, sine-wave mechanical oscillations around corners. This could easily explain 0.2mm extra width. Also blobs and overextrusion could explain that, if they would be present. Also, "elephant feet", the sagging of the first layers, could make a model seem wider than it is, if you measure it. Another option would be that calibration is off. But as gr5 said, it really draws its strokes inside of the model-edge, not centered on the edge. It takes its nozzle-width into account. Similar to image-editing programs where you can stroke a selection with settings: stroke inside edge / centered on edge / outside of edge. Here it is inside.
  19. You can always disable all brim in Cura, and design your own in CAD, exactly how and where you want it. I often do this when I want some features to have brim, and others not, or if I need irregular brim.
  20. I would say: start by marking all pulleys, belts, etc. with a marker. And see if they move relatively to each other. Or otherwise, check if the driver-chips don't get too hot (e.g. cooling fan not working, vents blocked), so they do a temporary thermal shutdown, until temp drops again to a safe level. Check if the cables to the steppers are okay, no bad contacts? These are the first things that come to mind, but I don't know your printer, so there could be other issues.
  21. Are you sure that nesting them reduces printing time? Printing the required support material for separation also takes time, and it increases the risk of failures, which also cost time. If you would use two glass plates, and preheat one while the other is printing, so you can switch glasses immediately after completion, it might go faster? Might be worth comparing both?
  22. I would at least try a couple in PLA. I have seen others who made things in PLA for emergency use too, recently. And I have also made prototypes in PLA for single use, to be desinfected with the usual 70% alcohol, or with some sort of chlorine solution (not sure about exact formula). 3D-printed tools are never optimal due to the pores, voids, and irregularities that collect "finger-mud". But this is for all materials, even PET and nylon. So you should make people aware that this is only an emergency solution, and that these pores are the draw-backs of this technology. If you would be familiar with ABS printing, you could do acetone-smoothing until it has a high-gloss. (I am not, so I can't give tips.) Acetone-smoothing also works on PLA (most of them), if printed in thin layers, but far less than on ABS. It tends to close some of the tiny gaps, but not bigger ones, and does not produce high-gloss surfaces. The only real disadvantage of PLA is that it is bio-degradable, but I don't know how fast and how much is it eaten by hospital bacteria, if any? At least, bacteria haven't eaten my sifts and tools in my lab, even not the always moist ones in the sewer. The advantages of PLA might outweight the disadvantages here: it prints fast, has good layer-bonding, good bed bonding, well known printing parameters, predictable results with very few failures. If you print in thin layers (0.06mm or 0.1mm) at slow speeds, it is watertight. If printed at higher speeds and higher layers (0.3mm), it is not watertight, in my experience.
  23. Hoi Anders, Do you also have data on *how long* a mask can be weared? Ours were chemical gas masks with filters depending on the industrial chemicals. As soon as the active stuff (e.g. active carbon) in the filter would run out, the filter had to be replaced. This was usually not very long: 30 min ... 1 hour, or so, depending on the filter, the gas concentration and the invironment. But that was 35 years ago, and products and rules may have changed a lot since then. I am particularly wondering on how long a medical mask can be weared before bacteria start to grow wildly? Their reproduction time is usually around 20 min, if I remember well. So, after 2 hours, you have 64x more bacteria than in the beginning, if they can reproduce in the filter. Also, I think a good mask should have two filters, both equipped with a one-way valve. One filter for incoming air, so others don't infect you. And the other filter for air you breathe out, so you don't infect others. If both flows go via one filter, and they all accumulate and start to grow, it might make matters worse in both directions. You inhale back your own bacteria, and you blow-out the ones accumulated from outside. Similar to blowing-out a dirty filter with compressed air. Then the filter is like carrying around your own legionella incubator. Sort of tropical CO2-enriched green-house, moist, warm. Also, a lot of medical filters do not stop (dry) viruses and bacteria, but they are designed to stop and absorb water droplets. This greatly reduces the risk of infection, but it still might cause a false sense of security, where people take greater and unnecessary risks. Home-made filters might work or not, very unpredictable. But if they are worn for days and days, without refreshing, then it might be much safer to put dirty socks over your nose...? I believe medical personnel should wear masks in these times, but they should take a fresh one for each patient, and even then replace it every 30 minutes or so... If worn incorrectly, filters might only work by means of the placebo-effect? People who are more confident and positive are less likely to get sick, compared to people who believe they will get sick? But this is just my view, based on common sense. I don't have scientific data to prove or disprove it.
  24. What does it do if you make wall thickness 0.41 or 0.42mm? When making small text, I make the legs 0.5mm wide instead of 0.4mm, to avoid gaps when the STL might make the walls just a little bit thinner than 0.4mm in corners (and cause gaps), due to the triangles. But I am using an older Cura version, so I don't know how the newer versions handle this.
  25. If you want to make videos (=editing, cutting, adding text, effects), a good freeware editor is "Shotcut". This is in heavy development and making fast progress. Almost as powerfull as professional editors, but currently not yet as stable: so you need to save projects regularly [Ctrl S] in case of a crash. It is very easy to learn and handy, and it does not eat too much computer resources. There are a lot of good tutorials on Youtube. Watch them first, to see if it appeals to you. For publishing of course you have all the Youtubes, Vimeos, and others. Just be aware that they may kill your account at any time, if you do not express the same politically correct view as them. I have seen lots of interesting documentary channels disappear the last year. So, always keep multiple backup copies of your videos locally.
×
×
  • Create New...