donmilne

What do you mean by "jam"? If you think the plate is too close to the nozzle then you should re-level, this time taking care to get the gap right (using e.g. paper as feeler gauge) at all three contact points. Instructions are in the user manual as I recall.

Perhaps I don't understand, and I'm on a tablet right now so can't test it, but two butt-jointed cubes would make an illegal shape if any internal faces remain. That would place the bug in your 3D design software, not in Cura.

+1 for Robert's feeder, though I actually use this variant of it because it straightens the filament too. You should print your preferred variant out ASAP, even if you have no intention of fitting it right away. Don't wait until the underextrusion is so serious that you no longer have the option to print it! I also like to use a printed filament guide to guide the filament away from the UM2 power cord. And Robert's filament cleaner widget. Judging from this thread, if Robert designed it, then probably you should print it.

Yes, I'm aware that you can do lost PLA casting - and I am kind of curious why Shiremog didn't, but it's the process for producing that wax model that I'm really asking about.

@shiremog: How did you produce the wax model? Did you 3D print using a wax filament? If so, where did you get the filament?

That looks like an excellent idea. And it looks like the snap on guide still fits, or is that modified too?

If we're talking about improvements to the forums my #1 priority would be making it easier to edit posts, ideally in plain text phpBB format since the "smart edit" feature consistently makes a mess of my post. #2 priority would be to make it easier (and possible!) to directly attach images and zip files to messages. Personally, I have little interest in forum seach - I always use Google anyway. I also have no interest in a rewards system, and IMHO messages should continue to be displayed in date order, rather than according to some dubious ranking system.

Large temperature errors could have several causes, with different fixes for each. There's no point in going into them until you verify that you have a problem, which IMHO is unlikely. And the +/- 1C thing should be taken with a large pinch of salt, what you're looking for is that the temperature is not wildly off. Temperature fluctuations can cause poor surface finish, but I don't see how it can cause underextrusion unless it not hot enough to melt the plastic. On the filament being quite hard to pull out when cleaning the nozzle. Without being there to judge what "hard" means, I would say that if true it could mean that the teflon spacer is deformed. Otherwise, a general difficulty removing the filament - if the Bowden is still in place - could mean that the Bowden is not pushed all the way in. That leaves a gap between Bowden and spacer in which a collar of plastic can form, making the filament hard to remove.

What is the filament? Was the printer working before? Why are you running so hot? Clean the head (atomic method), and the feeder.

I wouldn't trust a PC driven USB connection for as far as I could throw it. SDcard all the way.

Can someone tell me if the hotend PID changes are intended to influence surface quality as discussed in :- http://umforum.ultimaker.com/index.php?/topic/6613-firmware-surface-kwality-bug-1407-vs-1403 I appreciate that the other thread discusses the heated bed rather than the hotend, but there was speculation that the bed heater control was somehow affecting hotend temperature stability.

Aluminium in use is almost always an alloy, tailored for some characteristic (in fact the same can be said for every other industrial metal). If you search eBay for Aluminium rod there's a good chance that what you'll find is 6082-T6, it's probably the most common machinable grade.

I tend to think of PU as a curable resin. Can it be used as a thermoplastic? I can't find any details on melt temperatures etc.

Well, that is one of the things that makes SketchUp not a solid modelling tool. SketchUp will quite happily let you have objects which intersect, walls with 0 thickness, enclosure surfaces missing altogether etc. With great care you can avoid these defects, but you would have to have this in mind from the moment you started to the create the model. I do now try to make all my SketchUp models watertight, even though I only currently use SketchUp for woodworking designs, not 3D printing. p.s. Does your friend make use of the Outliner Window feature in SketchUp? I use it zealously, making sure that all model components are part of a named group or component. That way I can easily enable and disable drawing elements - and it should be easy to see objects that are in the outline but not obvious (e.g. because they are hidden inside something else).

Well, I would suggest that anyone who is especially sensitive should maybe avoid 3D printing. Sorry, but I know from experience that lots of very silly regulation arises out speculation like this. For example VOC regulations: one person coughed once after painting all day with the windows closed, so now we can't buy oil based paints.

How toxic can nanoparticles be? Their LD50 would have no be nano as well. This makes as much sense as homeopathy.

Sketchup is great, but it is not a solid modelling (3D manufacturing) tool. Creating solid models with SketchUp is possible, and easier with the pro version, but even then you need to be careful. Missing walls is a classic sign of a non-watertight 3D model. What kind of shape is it? You might find it easier to do it yourself. If the shape is simple I would use OpenSCAD. Yes, a picture of the model might be useful.

Well, (a) I don't understand what would be producing "nanoparticles": the PLA comes on a filament that is melted, not a dust. (b) PLA is biodegradable, so even if present I don't understand why they would be harmful. Hopefully the hypochondriacs and safety fascistas don't get to interfere with this hobby like they interfere my woodworking, metalworking, plastic casting... or just about anything else fun come to think of it.

I agree with George, and it should be easy enough to verify: just download any OpenGL test application and try it out.

I did say "usually" and "not ... complex". At best the object will lose all identity, it will just come back as a set of polygons. And, the shape can't have any attributes that STL doesn't know about. Depending on the precise format used there may also be precision problems introduced by using ASCII or single precision STL.

The clicking sound is the feeder motor failiing to turn to the next step, because the torque requirement is too high, because of high back pressure. That could definitely be caused by the bed partially blocking the nozzle, so don't dismiss that idea. It could also be be caused by a nozzle being blocked internally by something that doesn't want to melt (have you been using ABS? - anyway the Atomic method shld fix that), or a faulty temperature sensor not letting the hotend get hot enough. ColorFabb is a good brand, so it probably isn't the filament being too big, and a newish role too? So I doubt it's friction in the Bowden tube.

It looks like the bed is sloping slightly, the top bit (in the picture) is just a tad too close and the nozzle might even be bottoming out on the glass plate. If so this problem should correct itself on later layers - so don't be too quick to abort the print. I might also raise the nozzle temp by 5C to aid extrusion.

How does this affect the heating of the bed? I would have thought you have effectively added a large heatsink. Temperatures will be more stable, but harder to reach. Also potential burnout of bed heater.

I can't write a plugin for you, but I do suggest that you try tweaking the "minimum layer time" setting.

I already answered that for you in the other topic. CAD --> STL is a one way process. That fact is not changed by embedding the STL in a CAD drawing which you then write to a DWG file. Going from STL to CAD is a bit like trying to go from a paper book to a .DOC file. Scanning the book gives you a non-editable image format, which of course can be embedded within an editable document much as you did with the STL in the DWG. You would need something equivalent to a 3D Optical Character Recognition, but I doubt that such a thing exists.