geert_2

If you don't like using glue, because it is sometimes difficult to get the model off, or because you need to take the glass out of the printer for best results, then you might try the "salt method" first? In short: after cleaning, wipe the glass with a tissue moistened with salt water. (See link in my previous post for full manual.) For me this works great with PLA (Ultimaker) and PLA/PHA (colorFabb). It gives good bonding when the glass is hot (60°C), but no bonding at all when at room temp (20...25°C). So you can just take the models off, without any force, and without need to remove the glass from the printer. Then wipe again with a tissue, and you are ready for the next print. For me, the ease of applying the method without taking the glass out, the good bonding, and the ease of removing the models, are the main reasons for using this method. For some other brands of PLA this method works less good (although still better than nothing at all); and it does not work for ABS, PET, etc. So you will have to experiment for your materials and your circumstances. Stay around when printing the first objects, so see how well it works (or not) for you. It may be best to start with a small but difficult to print test model, with various bonding methods, and choose a method that works well and that you like. There is no perfect method for everything and everyone. And personal taste also plays an important role, nothing wrong with that.

To me it looks like in most photos the build plate is way too far from the nozzle, and the first layer is way too thick. I print my first layer at 0.2mm, and it is way thinner. Usually the underside of my prints looks like this, almost like a mirror. The camera often does not want to focus of the bottom surface, but focusses on the reflections mirrored in it instead. In the top photo the mirrored image is out of focus, in reality it appears much shinier. In the bottom photo the camera focussed on the reflections, here the cover of a desktop fan, instead of the bottom layer. But in real life we see both together. In your post I didn't see which bonding method you used, if any? If you would just print on bare glass, without any bonding, this might work very well in dry weather, but might completely fail in moist weather. At least, that is my experience. Also, bonding might completely fail after cleaning the glass with soap, with dish washer, solvents, etc. Soap reduces bonding, and solvents might contain traces of oils. So I clean the glass with pure warm tap water only. For bonding you could try gr5's method (10% white wood glue in water), neotko's method (hair spray), 3DLAC, the gluestick (maybe wipe with wet tissue afterwards), or my salt method (wipe the glass with a tissue moistened with salt water, see here: https://www.uantwerpen.be/nl/personeel/geert-keteleer/manuals/ ).

I would suggest: verify that your fans are set to 100% speed while printing PLA, and that they are running indeed. Then try printing this model slow (25mm/s?), cool (195°C?), and put a desktop fan at some distance (1...2m?) in front of the printer at slowest speed. Edges of sharp overhangs often have a tendency to curl up if printed hot and fast, on both of my UM2. This causes irregularities in the model, very similar to yours. The effect can be reduced by printing slow and cool, but not totally eliminated. If you don't want to waste too much material just for testing, try cutting off the bottom part of the model and only print the problematic top.

Forgot the STL-files for those who want to try it: overhangtest11e.stl (=table with supports) overhangtest12c.stl (=pipe with flanges) (Note: there is definitely a bug when inserting images: in my previous post, I did not include the last image. The system duplicated it itself. And I can't remove it anymore. Most of the time inserting images goes best when dragging and dropping them from the Windows explorer into the desired place in the text. But here one went wrong...)

Ja, sommige mensen smeren die lijm van de lijmstick nadien uit met een natte doek, tot er alleen een uiterst dunne laag overblijft, bijna onzichtbaar na drogen. User gr5 gebruikt een mengsel van 10% witte houtlijm in water, als bonding method. Je moet zijn film daarover eens opzoeken. User neotko gebruikt haarlak. Sommige mensen gebruiken 3DLAC (=lijkt op haarlak, maar speciaal voor 3D-prints). Ikzelf gebruik voor PLA alleen zout water: een doekje een beetje nat maken en daarmee de glasplaat afvegen, tot het droog is. Na afloop zie je alleen een soort "mist" op het glas, net zoals bij een wijnglas dat jarenlang in de kast gestaan heeft. Groot voordeel is: een goede hechting bij warme glasplaat (60°C), maar nul hechting bij koude plaat: dus de modellen komen vanzelf los, zonder geweld, en zonder dat ik de glasplaat uit de printer moet halen. Voor de volledige methode met beelden, zie mijn handleiding: https://www.uantwerpen.be/nl/personeel/geert-keteleer/manuals/ (En dan een beetje naar beneden scrollen.) Maar dit werkt alleen bij PLA, niet bij PET of andere materialen. Dan gebruik ik gr5's methode (10% witte houtlijm in water), maar soms lijmt dat zo goed dat het stukjes glas uit de plaat lostrekt. Ieder heeft zijn voorkeurmethode. Als de rest van je print goed is, zou ik voorlopig niet aan de andere settings prutsen. En nu eerst alleen proberen om je bonding en afregeling van de glasplaat te optimaliseren. Niet teveel ineens veranderen.

@smartavionics: based on your idea of a support-bridge below the real bridge, I did a few more experiments for single-sided overhangs, like roofs and flanges. This concept seems to work reasonably well for straight walls. It is better to have a few little defects in the shadow of the roof, than to have supports all the way down to the floor, and damaging all items in-between on the walls or the floors. I don't know the English terms, but in Dutch: vensterbanken, luifels, raamkozijnen, luiken, keldergaten,... In this test model I printed text on the floor, so I don't want to damage this text with supports. Keep in mind that the table-model is small: all walls = 1mm thick; stairs in the support = 1mm x 1mm; gap between vertical walls = 1mm. Connecting strands between model and support = 1mm long x 0.5mm wide x 0.2mm high. Gap between ribs on top and underside of table = 0.3mm. This works well on my UM2 for PLA, printed at 50mm/s, 210°C; or printed at 25mm/s, 195°C. For dual nozzle systems like the UM3, you could fill the gap between the ribs on top of the support, and the table-underside with PVA-support. The ribs would give a sort of dove-tail effect and improve bonding of PVA to PLA. At least, I would guess so. For circular walls like flanges on pipes, this method is not yet good enough, because when drawing the outline circles of the flange, the filament cuts the corners and draws a straight line instead of following the edges. Basic concept: supports connected to side walls via tiny strands. No connection on top (gap of 0.3mm). Gap between supports and side-walls is 1mm. Connections are 0.5mm wide x 0.2mm high. Don't want to destroy the text by supports going all the way down (for single nozzle printer). Same idea for flanges. But the supports need to be in segments, otherwise we can never get them off. Cut view. Results: colorFabb PLA/PHA (natural), printed at 210°C and 50mm/s, thus rather fast and hot for such a small model. The method should work in less than optimal conditions too. A little bit of damage to side walls after removing support, due to the connection strands. After a bit of cleaning with a sharp knife, the result seems pretty reasonable. And the table underside is okay, as good as it can get on a single nozzle printer. Remember: all walls are 1mm thick, nozzle = 0.4mm, text = 3.5mm character height, 0.5mm line-width, and 0.3mm raised. Removing supports under the flange. Flange itself is rather ugly and irregular. When printing the flange, the filament did cut the corners (straight line from hole to hole), instead of following the outline of the circle. So this created an irregular flange. Not good enough. Idea: for dual nozzle printers and PVA-support: when using a combination of support in PLA, with a buffer layer of PVA in-between, the connection between the PVA-support and PLA-support could always be done with such a dove-tail. This should make it bond well.

Yes, I have the no-save demo-version of moi. It is good if you know exactly what you want to make, thus if you have a clear picture in mind. And it can handle organic shapes too. But it is difficult to rework existing parts. So it is less suited for technical things that need frequent adjusting. It is more geared towards artwork. The good thing is that you can install a one-month free demo-version to try if it appeals to you, or you can install a timeless no-save version. It requires very little system resources, so runs on less powerfull hardware too. Also, it has a good forum and good support from the author.

I have no experience printing nylon. But from the specs it is known that in only 3 hours it can soak up enough water to make it unsuitable for injection moulding. So, the same will probably apply to 3D-printing. If you look at a printed nylon sausage under a microscope, you may see bubbles inside the material, or craters on the surface (=popped open bubbles). So using a method to keep it dry even while printing, is best. (One of the drying boxes available.) But if you are totally new to 3D-printing, I would recommend that you first print a few weeks in an easy material like PLA. Just to get to know the machine and the whole process. And only when you know these quite well, switch to a difficult material like nylon. It's like driving: you don't want your first driving lesson in rush hour in the center of Paris, in a big truck. You want a calm moment on a safe calm street, in an easy car. And gently progress to more difficult situations, as you master the easier ones.

In short: on the orange object, you need to move the *yellow ball* (=the reference anchor) of the arrows to the side you want to match, after which the yellow ball turns into a blue cube. Thus here you need to place it on the right outside of the orange item. Then you need to click the red move-arrow, select the "up to" function, and then click the left side of the green item. Then the selected side of the orange item will snap to the selected side of the green item. Thus: select the object(s) to move, select the source plane which has to match, select move direction arrow, select the "up to" function (this is probably what you missed), and select the destination plane to move the object upto. Here is the DesignSpark Mechanical tutorial on moving (Youtube): Be sure to study the tutorials too on Pulling, and Selecting, on the same Youtube-channel (you need to scroll down quite a bit): https://www.youtube.com/user/DesignSpark/videos?disable_polymer=1 I would recommend that you download all these tutorial (in high resolution 720p) and save them locally for future reference. In Firefox you can install add-ons like "video downloadhelper" to achieve this. I don't know for other browsers.

I looked it up in Wikipedia: "Gypsum is a soft sulfate mineral composed of calcium sulfate dihydrate, with the chemical formula CaSO4·2H2O." So it contains calcium, sulphate, and water. I am not a chemist, but I can imagine that this would interact with a copper sulphate solution. But bubbles might also be coming from the air trapped in the plaster: it is very porous, so it contains a lot of air. If you want to test which is the cause, you might put a little bit of the plating solution in different glass jars, and in one add water, in the other add plaster, in the third add primer paint (dried chunks), and in the fourth add PLA. And see which one gets destroyed.

Normaal zou de onderzijde bijna hoogglans moeten zijn. Bij mij wil de camera er soms niet op focussen, en dan focust hij op hetgeen erin gespiegeld wordt. In dit geval het beschermnetje en de bladen van een bureau-ventilator die erachter stond. Heb je al geprobeerd om de lijm dun uit te smeren met een natte doek? De nozzle iets dichter bij de glasplaat brengen zoals peggyb zegt, zou ook helpen. Er kunnen evtl. nog andere oorzaken zijn, vb. ongelijkmatig ingestelde printsnelheid van randen en infill, of onderextrusie, of filament met een onregelmatige diameter of samenstelling, maar best lijkt mij om eerst deze dingen te proberen.

Sounds like it scrapes something somewhere? If you would have a small mirror, such as a dental mirror or hobby mirror, maybe you could look behind it when moving the plate up and down? Or if you move it up and down manually with power off, do you feel anything?

Are Cura-settings "per user", or globally per computer? If per user, you could make a different username per main setting, like user: "Printer brand X - Filament material Y". First install everything as it should be (languages, keyboard layout, regional settings, all software, etc...), and then copy that profile to different users, and adapt Cura settings? I don't know if this is possible, I am just thinking aloud... Edit: the idea is that for every machine and material, you would need to log in as a different user. And then you would automatically get the optimal profile for that situation. For example username "PLA on UM3" would give you the optimised profile to print PLA on the UM3. And so on.

Also, when depositing molten filament in a circle, it is pulled inwards towards the center of the circle, before it has time to solidify. Like syrup would be pulled inwards. If the holes need to be accurate, a simple solution is to carefully go through them with a hand drill. No electric drills: they will melt the model if it is PLA. I bought a separate drill chuck for this, so I have a good grip on it, and I use that as manual drill. This gives enough power, and gives you a good feeling for what is going on.

Reasons for prefering DesignSpark Mechanical over SketchUp: 1) I never had any STL-file problems with DSM, thus never had printing problems. With SketchUp you are going to have them every time, and you are going to waste days and weeks, and get very frustrated if you have deadlines and work to do. 2) In DSM it is easy to round corners, or add chamfers, which is often needed in technical parts. 3) It is easy to undo things: to delete or change existing roundings on a design, move holes, change extrusions, etc... This is a major benefit when you often have to change existing designs. (Compare this to for example FreeCAD, another freeware 3D-editor: in FreeCAD you can create very complex designs, but most of the time it is impossible to change them afterwards, without breaking everything else.) 4) As gr5 said, in DSM we never have problems with "empty" cardboard models. It are true solids. Contrary to in SketchUp: very often the walls do not touch each other, but there are invisibly small openings, like a paper model house that is not glued together perfectly. It may look okay from a distance, but it is not water-tight. These are not solids, but "empty cardboard models" that can not be printed, and that are horrible to correct. 5) DesignSpark Mechanical is much easier to learn than for example Solidworks (although Solidworks is way way more powerfull). Especially if you come from SketchUp. 6) It is free, contrary to Solidworks and similar very expensive packages. Disadvantage is that you can not make colorfull renderings, so you can not add brick textures, wood, grass, etc... But these can not be printed on an FDM printer anyway, so no problem for me. If you need colorfull renderings and organic shapes, you might need Blender (free), or Maya, or something similar. For technical renderings, you might need Solidworks or similar. Or Form-Z. Also, DSM is somewhat limited in features, compared to its commercial sister SpaceClaim, or to Solidsworks. But for 3D-printing, it is more than sufficient. The best thing you could do is watch a few demo-videos on Youtube (also of other candidate programs, FreeCAD, Onshape, Blender, Solidworks, Maya, Form-Z...), and see wether it appeals to you. Not every user-interface appeals to everyone (for example, I can't find my way around in Onshape, while others can do wonderfull things with it). In the end, even if you spend a day watching demo videos, it will save you weeks of time. If you have a company and need to exchange designs with manufacturers, you might prefer a commercial package. As a hobbyist, or educator, you might prefer freeware like DesignSpark Mechanical. At least, this is my viewpoint, but of course you have the right to see things differently.

If the PVA (=white dissolvable support material) is clogging, and the underside of the roof will be invisible and unimportant, a temporary solution might be to print the roof with only one nozzle, and generate supports in the same material? And then break these supports off, just like we do on a single-nozzle printer. Of course this is not a permanent solution, but it might enable you to get the work done on the deadlines, for now. And in the mean time you can try to find out why the support material clogs, or dry it.

If the LED would be short-circuited, I could imagine that it would destroy its driver transistor. With the power off, you could try to measure with an ohm-meter on the LED, while in-circuit. This method is of course unreliable, but if you would get 0 ohm, then it is at least an indication, and then you might consider removing it. But it seems unlikely to me that the LED would be the cause? It rather seems like an overvoltage spike, causing all the damage, including destroying the LED? I read that a lot of switched mode power supplies cannot generate a stable output voltage if there is no load. Thus this applies to almost all modern supplies with tiny transformers running at high frequencies, instead of the old huge and heavy transformers running at 50Hz. These modern switched mode supplies need a minimum load to work well. So I could imagine that the output voltage would have been way too high, due to no load, and that your board got an overvoltage spike when you inserted the plug, and the overloaded capacitors discharged into your board? This is just guessing, since I don't know the circuits of the UM-board nor the supply, but it might be an explanation. Some of the better switched mode power supplies have provisions to switch off the output if there is no load. Or they have an internal resistor over the output, to always give it enough load to be stable (the old Delta Electronics supplies had this, around 1990). But maybe yours doesn't? But normally both the power supply and the printer board should have overvoltage protection diodes (transzorbs or zener diodes?), to absorb such spikes. I don't know if the UM board has these? Or maybe they weren't up to the charge? It is best to only connect such power supplies with the main power off, and the supply discharged before connecting. To discharge it, place a suitable lamp over the output, for example you could use a suitable car or truck lamp, and solder wires on it. (Not a LED, unless you put a suitable resistor and protection diode in series with it, or unless you put two LEDs in anti-parallel, and that in series with a suitable resistor.)

For those who want to install DesignSpark Mechanical in a VirtualBox virtual machine, I made a new topic: https://community.ultimaker.com/topic/22327-manual-how-to-install-designspark-mechanical-in-virtualbox/ You could try this if you do not have a Windows computer, or just for testing purposes, without influencing your current system too much.

Manual: how to install DesignSpark Mechanical in VirtualBox DesignSpark Mechanical ("DSM") is freeware 3D-editor. It is a limited version of the commercial SpaceClaim program, and distributed for free by the well-known elektronics supplier "RS Components". It only requires registration to activate it. DesignSpark Mechanical is very well suited for geometric shapes, technical models and machine parts, like the one you can see in this screendump. It is not suitable for organic shapes like animals, plants, etc. It has its own fileformat "RSDOC", and it can export to STL and OBJ for 3D-printing, and JPG for images of the model (these are basically screendumps of the current drawing window). DSM can also read STEP-files, but can not export it, unless you buy the optional IGES/STEP import/export add-on. DSM is very easy to learn: it has a similar simple push-pull user-interface like SketchUp. But it creates very good, solid STL-files for 3D-printing. In 2 years time, and after hundreds of models, I never had any problem printing its STL-files. (This is contrary to SketchUp, which always causes huge trouble when 3D-printing, since it was never intended for that. SketchUp was ment to create "cardboard" 3D-images of architectural buildings for Google Earth, before Google had its own program to automatically convert aerial photos into 3D-models. So don't use SketchUp for 3D-printing. Use it for what is was designed for: making nice images of buildings.) On Youtube and on other websites you can find a lot of good tutorials and demo-videos on how to use DesignSpark Mechanical. You could also use the SpaceClaim tutorials, if you keep in mind that DSM does only have a limited set of the features of SpaceClaim. One disadvantage: it appears that DesignSpark Mechanical (and SpaceClaim) only exist for Windows. However, you can install it in a VirtualBox virtual machine. This might be a solution for Mac and Linux users. If you can find an old Windows 7 installer-DVD from a scrapped computer (to keep it legal), you could try installing Windows in VirtualBox. And then you can install DesignSpark Mechanical in that VM. The following manuals describe how to enable 3D acceleration in VirtualBox, and how to install DesignSpark Mechanical in a virtual machine. It requires a bit of experience with software-installations. The first manual is "printed to PDF-file" from the internet. The second one is my workflow, which I wrote down for future reference. virtualbox_how_to_enable_3d_acceleration_1.pdf install_dsm_in_virtualbox_1.pdf

Yes, I think so. DSM is derived from the commercial SpaceClaim: it has the same user-interface, only it has less features. (So you could use the SpaceClaim manuals too, as long as you calculate in that DSM does not have all SpaceClaim features.) But DesignSpark Mechanical can run in a VirtualBox VM, although slower and at lower rendering resolution. So, for those with a Mac or Linux, they could try to install it in VirtualBox, if they really want it. Find an old genuine Windows installer-DVD from a scrapped computer (so you stay legal), and try that. You need to google a bit on how to enable 3D graphics in VirtualBox. I don't think I can insert PDF-files here in the forum? Otherwise I could add the manual I wrote for my own reference, on how to install DSM in VirtualBox. I could copy the plain text as well if you want, but it is quite long. Or maybe I should make a new thread for that?

Here the print results of my free hanging support tests shown above. (To avoid confusion, keep in mind that this is a manually created test model, not automatic support generation.) The basic idea of a support bridge to support the real bridge works great. The underside of the real bridge is as good as it can get with a single nozzle printer (UM2). And the support is easy to remove. So I am definitely going to include this concept in my inventory of support methods. Thanks to smartavionics for the great idea. But my model still needs refinements. The triangle support shapes do need modifying. This will take some trial and error. Now the edges curl up way too much at the outside (more than 1mm, at 0.1mm layer height). Then the nozzle crashes into these curls violently, and tends to knock the model off the build plate, or to separate the supports from the model. As you can see in the light green model (PLA, 210°C, 50mm/s). The dark green model survived, but was printed much slower (25mm/s), so the nozzle crashes were not as severe. Also, for optimal stability the free hanging supports would need a few little strands to connect them to the main model. Relying on the spaghetti works well for the first layers, but after a couple of layers there is no more spaghetti. So there it needs a few connecting strands. Using strands in well defined places might also reduce the amount of damage to the side walls of the real model. This also will take some testing. But the basic concept definitely can be used for printing small overhanging roofs and flanges.

The only real solution is to not use SketchUp for 3D-printing. It is going to keep causing you endless trouble with every model, and it is going to slow you down a lot, and cause endless frustration. Try DesignSpark Mechanical instead. This is freeware, it only requires registering. In two years time, and many hundreds of models, I never had any issues with its STL-files. Never. DSM is very easy to learn: it has a similar push-pull interface like SketchUp, and there are lots of good tutorials en demo-videos on Youtube and elsewhere.

Actually, I use 14.09 (so I entered 15.04 as closest nearby). The reason: it works well for my old UM2 (non-plus). It is like when you have an old chainsaw that still works, then you just keep using it, unless you specifically need any new features. Also, last year I had to make lots of prototypes on deadlines, so I didn't have too much time to experiment, and I wanted to keep my system stable. But I know it is a bit overdue, so this year I am really going to update (install the new version parallel to the old), and I am going to try out all the new fascinating features. I think since a lot of the improvements are specifically for the UM3 and dual nozzle printing, it is indeed logical that they follow the updates more closely.

I have never tried printing "images", so this is guessing. Cura should be able to print JPG-images. So if you save your design from GIMP as a JPG-file, it should work. I don't know if black is higher and white lower, or vice-versa. If the background also gets printed, maybe you could lower the model a little bit into the build plate, or cut off the bottom layers, so this background disappears?

Since I am not related to the Ultimaker company (I am just a customer using UM2 printers, no employee), I can not give any official recommendations. I just used common sense and trial and error for this. Petrochemical oils and solvents often damage rubber, so they are to be avoided. But pure silicone grease is chemically quite inert, so I figured this would do the least damage, if any. I don't remember which brand I used, because I have several of those tubes laying around here, some still in their original packaging, some not (now in a glass jar). I applied it very sparingly, only a tiny little bit on the very edge of the rubber, where it scratched the flange of the pulley. But definitely not on the teeth, so they would not slip and not jump teeth. If you would like to try this method too, I think you should ask for: "inert thick white silicone grease, like used in binoculars and microscopes, and it must be rubber-safe". Most bigger hardware shops have it. Auto-parts shops also have it here in Belgium, because this is also used for wiping the car door-rubbers in winter against freezing up. But probably your shops are going to have different brands anyway than here. I haven't had any squeeking noises since, and the belts are still alive.