gr5

cura is always fast for me. I suspect your model has more polygons than you need. I find running it through meshlab as explained in the following article can speed slicing up a lot. If you follow the steps below, when you load the model it says how many polygons/triangles are loaded. Try reducing that by 10X to see if the model still looks nice and clear/sharp. If so it should slice 10X faster hopefully. http://www.shapeways.com/tutorials/polygon_reduction_with_meshlab

Not yet. But there is mesh mixer which lets you click on a spot to add a support and then click and drag to connect supports and delete supports and so on. It makes it pretty easy to do minimal, effective support: http://www.extrudable.me/2013/12/28/meshmixer-2-0-best-newcomer-in-a-supporting-role/

No. Just try it. However it could be quite tedious to put in 30 tweaks at 30 different heights simply changing the nozzle temp on each one. There used to be a plugin for cura 15.* which you might want to use instead that is designed specifically to change the temp randomly. I recommend installing cura 15.* and locating that plugin instead. But it will be work - you'll have to learn a new (simpler) cura and locate the right plugin, download the plugin, put it in the right folder. From that point on it's easy.

It will be much better I'm sure. But possibly not good enough if you want it well balanced. So you may have to come up with a procedure to adjust the balance or to machine it a little such that it is more round.

I don't think you can control fiber orientation as it is chopped up into tiny pieces and mixed in with the filament and the fibers are at random orientation. Is this Nylon with CF or is this PLA with CF (or maybe there is another material. Because the printing pattern should not matter. But it does if you have underextrusion or if you are printing a higher temp material like Nylon and you aren't getting good bonding. So let me know and I can help you with that. I've done lots of tests of material strength. I have a machine that tests modulus (a flexibility test) and ultimate strength. I have tested parts in different orientations. when you do it right (we can get into that later), it doesn't matter what direction you print. However if you have a mark forged printer, the fibers are LONG and it definitely matters. but Mark Forged has it's own slicer so I assume you don't have a Mark Forged.

Well before printing her, look at the part in layer view with support turned on and off and think about what will happen when it prints.

I think the above link has one of your answers. I don't know about the answer to your support question. I mean how would you get the support out? Are you talking about PVA disolvable support?

Okay so your issue is very unclear but I'm pretty sure I got it now. I downloaded the STL you are talking about. What you want to do is check the feature in Cura "ignore small z gaps". This will fix the issue you are talking about at the expense of taking 5% longer to slice. I think this should be checked by default but it isn't. Basically there must be some kind of pre-processor that looks at the slices and sees that in e.g. 5 layers a top surface is coming up and settings say we should start solid infill now - but it doesn't realize that before then there is also a bottom surface coming up. Or vice-versa. Anyway check this box to fix your issue.

This forum is new and we are still trying to figure out the moderation stuff although so far your first comment was approved within I think an hour and definitely the one above within 10 mintues. More the problem might be that the cura software people are taking the weekend off (most likely). After you have made 10 posts moderators shouldn't have to approve you. Feel free to make 7 more posts of one word each and I'll approve them all so you don't need to wait in the future.

I can't tell in the photo if the red is any better or worse but what you describe - "dragging through" - on the bottom layer sounds like you are indeed "close". There is a thin line - or not even a line - there is a tradeoff between too close and too far. I personally prefer "too close" because the more you squish that bottom layer the better it sticks to the glass. I mean I can print a UM robot and literally pick up the entire printer by that little robot. But for smaller objects (say under 70mm in x,y) that have a large surface area (say as opposed to printing a table where only the 4 legs touch the glass) you probably don't need it to stick as well to the glass. So for some prints I prefer it too close. And other prints I prefer it a little farther. Just so you know what to look for. If it's too close the bottom layer will be transparent. If it's too far the bottom layer will be ropy (like a rope). You want it somewhere in between. 90% of my prints are not meant to be pretty so I don't really care how it looks so I want it close. I'd really rather not have any chance of the print coming loose.

1000 hours printing!!! Yes! change the teflon. It needs changing more often than that. The price from UM (and other resellers) is quite reasonable! But don't get one from China as the teflon from UM and from 3dsolex is much better.

Nice! Great news!

Then check that belts are equal tightness (including the short belts on the steppers!!!) And look straight down on the printer and see if the rods through the head appear to be perpendicular.

Anyone who reads the above post FIRST FOLLOW HIS LINK TO CARIFY! Well this part is very different than the other part which had sloped sides. This part seems to have mostly vertical sides so it seems safe to increase top/bottom thickness for this part. You are going to learn that you have to change slicer settings for different parts for different reasons (otherwise there would be no need for changing the features). So: 1) It seems like you can increase top/bottom thickness for this particular part. 2) Why did you want to get rid of it in the last part anyway? Were you just trying to save time? I mean it's invisible inside the print anyway, right? Or was it ugly when people turned the part over and looked from the bottom into the infill? This part in this topic though is solid so why do you care about the extra infill? It *does* have a purpose - basically it is supporting the future shells coming up in a few layers time. Maybe I don't understand what you mean by "top internal layers".

Can you be more specific? Do you mean when you switch to ABS the default temp stays at 210C in cura?

Oh, okay. Good to know. I have only ever used octoprint for one printer and only printed maybe 1% of my prints using octoprint.

You can get something similar (but a little different) by setting the shell width to basically infinite. Set it to 200mm. That way there will be no infill and the shell will be so thick it connects the outer surface to the inner surface. HOWEVER, it won't be exactly like what you see above. Maybe you should be a little more specific. There are other tricks to get what you see above. You can create inner structures inside the part that force cura to do this kind of pattern. I guess I need to more. Is your goal: 1) to create microchannels for air/liquid to pass through? 2) To control grain/stress direction? 3) To make the part weak in certain areas such that it can be broken apart into specific shapes? Like an adjustable shim? 4) Other? Do they need to reach the left and right sides or is it okay if the pattern turns 90 degrees when it gets close to the right and left sides?

They care but I think they care more about paying customers who are screaming/begging for hundreds of other features and definitely not for this feature since it isn't supported by any of their current printers. This feature helps Ultimaker's competitors, and not Ultimaker itself because the UM3 doesn't support octoprint at all (I'm pretty sure) and the UM2 will work with octoprint but it's not officially supported (printing through USB is not officially supported on UM2). So I think they are actually very nice to release this feature in Cura considering they get little in return. Anyway don't you think it's great that you actually got a response from the guy who wrote the feature? That's pretty impressive right there, right?

It's true my first comments didn't answer your question directly but still I tried to be helpful. You didn't answer the question about if you got octoprint to connect to your printer already as if you can't get that to work then there isn't really any point to get cura to connect to octoprint, right? Anyway it sounds like @ahoeben gave you everything you need, right? You can ignore the instance it sounds like and just give it the host and port. Or is there more? [okay - graying about the below paragraph because it's completely wrong] One copy of octoprint can support many printers at the same time. This was kind of the original purpose - originally you could have many printers all connected to one raspberry pi (or in your case one computer) through USB and octo print could control all of these printers at once. I think the idea was, like an octopus with 8 arms, it could control 8 printers at once (or maybe much more than 8, not sure). So this would explain why cura needs to know more than just host and port? maybe? It would also need to know which of the (possibly many) printers octoprint is connected to. So if cura needs more than host/port then maybe it's which printer and try maybe 0 or 1 (the only printer - again you didn't say much about your setup yet so I'm not sure but guessing you only have one printer hooked up to octoprint). Also it might be that cura doesn't support non numeric host names such as the suggested "localhost". So alternatively you could try "127.0.0.1" for the host name instead of "localhost". Don't include any quotes in the host name.

I think the glass got worse, not better. At least for me personally the glass on my UM2 and UM2 extended (which are a year older) is much flatter than the glass on my UM3. In fact, early on, I was testing it on a table and noticed it really did not take much force to level (bend) the glass (a few ounces). So I fixed the problem by bending up the rear two corners of my print bed. I really pulled HARD on that aluminum bed to get it to bend up. Only later did I realize I risked breaking the trace on my heated bed and destroying it (oops! )

What kind of printer? It's the printer, not the slicer. Make sure the belts are of equal tension in X and Y and maybe calibrate and also make sure the X and Y axes are perpendicular. The most common problem on a UM printer would be that the two rods going through the head are not perpendicular. If it's a UM printer then push the head to the 4 walls and make sure the blocks hit at the same time. If not loosen some of the set screws and re align and then tighten the hell out of those set screws and double check. The symptom would be errors in the diagonal direction - in otherwords bottom left of the desk to top right would be a different diameter than the perpendicular measurement. If X and Y are not exactly equal on your printer then you would get different distance in those axes instead. Another common problem on the old UM originals was the pulleys were not colimated - in other words the hole through the pulley was not through it's center. This results in for example if the head is supposed to be moving at a constant speed it instead speeds up and slows down each time the pulley rotates once. But you should also consider putting a shaft or hole in your disk so you can mount it on a drill or lathe and file/sand it down a bit while it's spinning fast to get it better. *and* also consider adding bits of glue or other weights to make it well balanced. I took apart an old cassette player recently and was amazed at the heavy round part that was connected to the capstan (the primary speed control shaft). Even though it spun kind of slow it had all kinds of last minute adjustments to make it perfectly balanced. And it was clearly made very accurately on a lathe at some factory yet they still had to drill out a few chunks to balance it.

I don't think this part needs any support. Or at least enable support only touching buildplate as conny_g says. Personally I rarely use cura support and instead generate the support myself in cad. I would create two arches in the front and back of that large cylinder and do zero support in between (let it bridge across the arches.

tristan I think didn't look carefully - you want your travel sped high - on an ultimaker at least 150mm/sec and 30mm/sec is definitely slow. A good speed for good quality. Are *all* the cura printing speeds the same though? Sometimes you can set it to 30mm/sec but the infill speed is faster and that will cause these over extrusion blobs on harley. I would also increase layer thickness to 0.1mm but I'm not sure that is the issue. Did you have support enabled for Harley? that could explain the entire problem. It might be as simple as disabling support.

I suspect the main problem for the 3 people here who have seen it is the glass. The glass that Ultimaker sells is tempered and tempered glass tends to be thicker in the middle. I don't know why they use tempered glass. I think regular glass might be better. It could be a very carefully thought out reasoning with lots of testing -- or it could be just that "tempered" sounds better. I mean as an engineer I would definitely not have considered tempered - instead I would have considered pyrex which has a much lower expansion coefficient. But tempered and non-tempered regular glass have the same expansion coefficient and need to be allowed to expand and shrink (hence the clips that hold it in which allow it to expand slightly. People on the forums have gone this route without problems. Again - I think it was a mistake to go to tempered glass but maybe there's something I don't know. Anyway there are 2 types of glass that are probably superior to tempered glass for 3d printing. One is regular glass - the most common type - found at any glass store - and glass stores are found in almost every town in the world - at least any town that has lots of houses in it with glass windows. So anywhere you see glass windows in your area there should be a local glass shop that can sell you cut glass to fit your bed and they can even grind the edges so it's not sharp. All very cheaply. The other type of glass is called "neoceram". Or ceramic glass. It is used for high temperature settings - mainly fireplace glass. I know some ultimaker owners who bought neoceram glass and were extremely happy with it. It is usually much flatter than tempered glass but not always - some people who bought neoceram said it was flatter than .01mm edge to edge and others got pieces that were 0.5mm of thickness variations. So who knows. I've been thinking I should sell UM replacement glass but don't have the motivation right now to experiment.

In general I would expect the print head to work better if the print wobbles. It's definitely true that if you have two printers on the same table and the table is wobbling from the printers then the prints (or portions of prints) created when both are running have a lower quality. But if the only thing wobbling the table is a single printer I think it should be fine.