Jump to content

geert_2

Ambassador
  • Posts

    2,084
  • Joined

  • Last visited

  • Days Won

    34

Everything posted by geert_2

  1. Yes that seems like an excellent idea, if it can be done without too much damage to the side walls (in single nozzle printers with only one material). The concept might also work for small overhangs like roofs, if you would build a sort of triangular support that is only sticking lightly to the side walls, just enough to keep the roof in place, but still easy to remove. I have quickly modeled such a thing in DesignSpark Mechanical. It's a very crude model, just to try the concept. The supports (yellow and red) are totally separated from the model (cyan). For adhesion, they would rely on the spaghetti and nozzle leaking that is produced in the first layers of these supports. Currently the horizontal gap is 0.5mm, and the vertical gap between the ribs and roof is 0.3mm, but these might need adjustment. I am curious if it is going to work? Next week I will print it (I am not here tomorrow). Edit: dimensions: center overhang = 30mm, side overhangs = 5mm, width = 15mm. Little ribs on supports = 0.5mm wide, with gaps of 1.0mm in-between.
  2. Yes, I understand. But a lot of other people don't know. So that is why I needed a demo-model to show them the effect with and without supports. But for such a single-sided overhang like the bow above, if printed without supports, I have been wondering if the results would be better (=less spaghetti, but of course still some spaghetti) when the first layer is printed length-wise? Or if it is printed transversal, line by line, maybe with a reduced line-spacing (=thus with some overlapping), so that each line still glues a bit to the previous line? Very hard to predict. In real life, printing small horizontal overhangs might be usefull for an architect who wants to print a house model, and the roof is overhanging a little bit from the walls. Or if you want to print a vertical pipe with a little flange. For my real designs, usually I include the supports in the CAD design itself, like this (=the 2 purple items with ribs), for printing on my UM2. The hollow pink block is a dummy to provide enough cooling while printing the tiny top area of the model.
  3. Last week I designed a little test model to demonstrate the limitations of an FDM-printer to people who have no experience with it, but who want something designed and 3D-printed. Purpose was to show big overhangs and tiny extensions that get not enough cooling (the little cones at the top). I also used it to experiment with temperature and speed settings, for best results. But the model might also be usefull for this new bridging function. So, feel free to experiment with it. Size is fairly small: ca. 40mm x 30mm x 30mm, about 1.5h printing time at 50mm/s. The item at the back is a real bridge, with at one side a small horizontal overhanging plate, and at the other side a 45° overhang. The bow at the front only has support on one side. So the goal here is to find settings that do produce the minimum amount of spaghetti, and the minimum of curling upwards while printing. STL-file: overhangtest3d.stl
  4. I don't know the UM3, so I am just guessing here... Since those items are printed very close to the edge of the bed, could it be that the print head hits the back somewhere, or it hits something else? And this blocks it from going further, and thus causes it to skip steps? If you print a little test model in the center of the bed, let's say a cube of 10mm x 10mm x 10mm, without anything else, does it also have this offset?
  5. Yours is not a standard UM2 nozzle. The UM2 (non-plus) does not have a removable hex nozzle, see the pic below. (Don't mind the wiping wire, that is my add-on.) After cleaning the nozzle with an atomic pull, you should be able to see through it from above. You could try heating up the nozzle manually to 210°C, and manually feed some filament through it. It should extrude easily. Then remove the bowden tube at the feeder side, and manually feed some filament through the tube. This too should go easy. In this way you could manually check the whole feeding traject for too high friction. For a more gentle atomic pull, see my manual at: https://www.uantwerpen.be/nl/personeel/geert-keteleer/manuals/ It also contains a few other tips on nozzle cleaning. Also, if the filament is near the end of the spool, it is wound very tight. This will cause very high friction in the whole feeding traject: in the bowden tube, but also in the nozzle. It will also resist unwinding, like a strong spring. Especially if it is old PLA that has become hard. To solve this, what I do near the end of the spool, is manually unwind some filament, wind it up in the opposite direction around a skater wheel (7cm diameter), release it. Then take the next part, wind in the opposite direction on the skater wheel, etc... This straightens it. And then I let it wind up again on the spool, but now it will sit very loose, with a bending radius of ca. 30cm. It will no longer cause much friction.
  6. I have a very old Dremel (ca. 1980?) without speed control, so it always ran at 28000 rpm. But you could buy an external speed control back then. Originally this went from ca. 10000 rpm to 28000 rpm. But I opened it and changed the settings of the potentiometers (=variable resistors), so now it goes from zero to 28000 rpm. One big disadvantage: at low rpm the Dremel has not enough cooling, since this is provided by the internal fan which does not run fast enough at low rpm. So it gets very hot and requires frequent cooling pauses. Also, at low rpm it has very little torque, but that may be an advantage here: then it is less likely to cause damage when getting stuck. Google for the words: dremel solid state motor speed control And then view the images. Be careful if you buy a second hand one, since a lot are for 110V AC, and most have an American plug (even the 220V models). Or maybe you could buy a simple universal motor speed control? Put the internal speed control of your Dremel to maximum, and place the universal speed control in the 230V mains cable. This will probably be a lot cheaper. Just expect the neighbours to complain about "snow" on their TV, or noise in the radio. Google for: 230V universal motor speed control
  7. @peggyb: I have seen these in Youtube videos, but I never saw them in real life. Are these brushes ment for metals? Or would they work well on plastics too, without melting them, and without cutting through them (of course at low speeds)? Do you have any experience with this?
  8. If I had to print this, I would try it upright. If it would be too long to fit in the printer, I would cut it in half, and add a sort of flanges to make glueing together easier, like in this quick 5 minute sketch (made in DesignSpark Mechanical). And maybe I would also add a little brim to prevent it from being knocked over, depending on the height/width ratio, the material, and the bonding. I think this would give a much better quality.
  9. In itself there is no problem with more columns. The problem would be if the data is divided over multiple columns, for example: Column A [TAB] Column B [TAB] Column C Nozzle temp: [TAB] °C [TAB] 210 Number of skirt lines: [TAB] 3 Here the temp is sitting in column C, while the number of skirt lines is in column B. Then you can not simply copy and paste the whole column with data into an Excel worksheet, into a prepared template. It would require manually re-arranging some data into different columns. For printing as a plain text file, it does not matter of course. The idea is similar to when copying data from a TXT-file into an Excel worksheet, to produce a graph. You want all the X-axis data in one column, and all Y-axis data in another column. If this would be a material test that is repeated 100 times, then we can simply copy the column with X-values into a prepared worksheet, and copy the Y-values, and immediately have a nice graph. It wouldn't matter if you would use two columns, like this: Description (units) | Numerical data Or three columns: Description | Units | Numerical data Or whatever else, as long as all items use the same layout, and all numerical data end up in the same column. That was the idea.
  10. If you want to stack items very close together, you can always multiply and arrange them in your CAD-editor, and save that as a new design. For example, in this way you can place items in a ring (otherwise the area in the ring would be wasted), or in some other hollow area. Or you can place items in-between other items with fingers, and so on. For your wheels, you could place them in a very tight hexagonal pattern. The printer will consider it as one design, and print all together.
  11. Concerning your photography setup: what about modeling and 3D-printing a stand to hold your lights? I dislike hand-polishing too. So, for the polishing, have you tried a Dremel tool with very soft polishing disks? Or dental polishing disks or cones? They come in very fine shapes too. I guess it would have to be some tool that runs a lot slower than a Dremel, and a very wet polishing disk, to avoid overheating and melting of the PLA. For example the kind of wet polishing paste and slow running motor in the first video, combined with the fine tools in the beginning of the second video? (But definitely not the speed in the second video, that would melt everything.)
  12. I have also had it once when printing PET and using glue. I didn't use force to remove the model. And I heard a weird cracking sound while it was cooling, so I guess the glass chipped while cooling, even before I touched it. For PLA I only use my "salt method" for bonding (=wipe the glass with a tissue moistened with salt water, nothing else). This never causes issues, but it only works for PLA. For other materials where bonding is required, I guess you will occasionally have this problem. Mainly for this reason, I think a ceramic build plate might be beneficial? Or try a different type of glass?
  13. Addition: For easy importing in Excel, for example if you want lots of profiles in one big table, for a good overview and easy comparing, it is desirable to have only two columns (A and B) in the text-output. And the units being part of column A. Like this: Column A [TAB] Column B ... Profile name: [TAB] Normal PLA User comment: [TAB] This is my usual PLA profile Printing speed (mm/s): [TAB] 50 Traveling speed (mm/s): [TAB] 150 Nozzle temp (°C): [TAB] 210 Number of skirt lines (#): [TAB] 3 ... In which [TAB] is of course a tab-stop. A semicolon would also do (or better, user-selectable). But no comma's, since they might interfere with European decimal points. In this way we can open such a file in Excel, and directly copy column B into a template with a big table. Or we can easily print it from Notepad, or from Word.
  14. If the stepper motor axis is freely rotating in the pulley, I think yes. If the stepper motor isn't rotating at all anymore, I guess there is something else. But I have no experience adjusting belts and pulleys. So I would rather get a second opinion from specialists like gr5.
  15. Very simple setup indeed. But very effective, since the results often look like professional studio photography.
  16. Ah, I see. I don't know the electronics, so no direct answer here. But I seem to vaguely remember: wasn't there some function or some setting to switch off the white LED lamps when the print finisched? Or some gcode to achieve this (in which case it should be entered into every file, or automatically appended)? Maybe you could find something along this line? Anyway, after the print finishes, you also need to make sure the printer is on long enough to cool the nozzle, so the heat does not travel upwards into the filament, and does not melt and deform it above the nozzle, and cause trouble for the next print.
  17. In your first video it looks like you are rotating in the wrong direction? Thus loosening the screw instead of fastening it? Could you have a look at that? Normally the cork screw rule applies: turn right to fasten, left to loosen. And unless the video is mirrored, it appears you are turning left? In hard to access areas, or when doing it from the back, or out of view, it may not always be obvious in which direction to turn. Then just take out your cork screw and hold it in the same position.
  18. I would also recommend switching to DesignSpark Mechanical. It is freeware; you only have to register, which is a very fair price for the value that you get. DSM has the same push-pull style of user interface as SketchUp, so it is easy to switch. And there are tons of good tutorials on Youtube. You can learn DSM in less time than it takes to solve this issue. I have been using it for ca. 3 years now, and *never* had any problems with its STL-files, even though I made hundreds of models.
  19. For this sort of models, I would recommend that you have a look at the freeware 3D-editor DesignSpark Mechanical. This is very easy to learn (100x easier than Blender), and very well suited for geometric models like this (not for organic shapes). You can watch a few demo videos or tutorials in Youtube, to see if it appeals to you.
  20. This is a good idea, worth looking into. I am not a programmer, so I am thinking purely philosophically too here. If the print-routine would take the output to the screen as its input, and format that into a nice TXT or CSV file, wouldn't that work? Then you have a nice list of all relevant settings, and you can print, review, compare and manually enter them again if ever required. Then this print-routine wouldn't have to be aware of the internal structure of Cura? Because the merging of profile defaults, custom settings, and whatever else, is already done at that point? Of course the routine can't just use the pixel-stream to the screen. It has to intercept the communication somewhere earlier when the text-content is sent to the GUI-system, before it is converted into pixels. Or am I missing something? For me a simple TXT file would do. So that you can open it in Notepad, Notepad++, Wordpad, Word, Excel, or whatever editor you like. And you can then easily format it and write to PDF if required, or print on paper. This would be the most versatile and probably easiest to implement. With such TXT-files you could also input multiple TXT-files in Excel in different columns and then compare them next to each other, and print tables if you want.
  21. In my experience, minimum layer time causes other issues: this reduces the flow rate, so the plastic stays longer in the nozzle and thus becomes hotter and more liquid, which to some degree tends to destroy the desired effect: getting more cooling. The melt gets more liquid, instead of less. Try printing cooler and using the desktop fan first. On an UM3 there may also be issues if there is too much distance between support and model, so the model does not glue well to the support. There has been a recent thread on this and how to solve it (if that would be a cause in your model). But I can not comment on this issua, as I don't have any experience with UM3.
  22. Webcam? This is probably the easiest to install, and you get a lot more feedback from it than from a single signal. If you take a webcam with motion-detection and an output relay (e.g. to trigger an alarm, or to switch on a light if there is any motion) theb you can use that if needed.
  23. For this sort of models, you might want to try DesingSpark Mechanical. It is very good at geometric shapes. Never had any problem with STL-export and 3D-printing. Also, it is as easy to learn as SketchUp. To get an idea of the user-interface, and if it appeals to you, try a few demo-videos and tutorials on Youtube.
  24. On my UM2 (=single nozzle) I also have this when printing overhanging edges. They do curl up as they cool. Overhangs are not very stable anyway, since half of the extruded sausage is printed in the air, without something to support it (at least in single nozzle printers like mine). So, as long as it is flexible, it is likely to curl, in this case curl upwards due to shrinking. Small models suffer from this more than big models, due to less cooling time per layer. The best solution for me was to provide way more cooling: put a desktop fan of 30cm diameter in front of the printer, and set it to its lowest speed. So that the heat is evacuated fast enough from the print. But not so much that the nozzle and bed have problems keeping their temperature. Of course, printing as cool as possible (and thus also slower) also helps, since that makes the new layer cool faster, and it is less liquid when deposited.
  25. This seems to be caused as follows: while the nozzle travels over a gap, it still leaks a little bit. So there is a little droplet hanging on the nozzle. As soon as it meets the next wall, that droplet is deposited against the wall. It is sort of wiped off, and now sitting on the outside of the wall. The next time, the new droplet is deposited onto the previous droplet, since that is the first thing the nozzle encounters in its travel. Next, the new droplet is deposited on that old one. And so on. This causes these strange upwards directed "insect antennas". Printing cooler and slower reduces the effect, since the nozzle does leak less, and pressure in the nozzle is less. But it depends a lot on the material too: in PET I have this a lot more than in PLA, in which it is very rare.
×
×
  • Create New...