jenschr

@gr5 There's no unsupported layers. These are the models I am printing http://www.youmagine.com/designs/thin-components-for-raaco-storage The infill is on the print-bed, so nothing hangs. @illuminarti Yeah. I'm sure that @Daid is right about the calculations being correct, but it still does not look correct. All I want is to have parts that are as transparent as possible and I want to be able to do this with Cura since it's my favorite slicer

@illuminarti I only have the last one (as I typically just overwrite on the SD), but this is the one with 0.8 outline and least transparency: http://flashgamer.com/a/infill/RoundedTestCube_0.8.gcode @IRobertI Please have a second look at the last picture I posted. Same pattern, just much more transparent on the box to the left. It is really strange to me...

Right... So - how do I get "100% infill"? What is wrong with the settings I posted - or is it something else? http://flashgamer.com/a/infill/colorfabb_xt2.ini Test file posted here: https://www.youmagine.com/designs/rounded-square Here's an image of this box next to another box I made back in August 2013, probably using Kisslicer but it might have been Cura also. Notice how you can see the internal divider walls of the Kissliced one? I happened to jot down the settings I used as well on the box itself, so 50mm/s at 100% infill.

Ok. I feel like I'm not being believed here, so I've made a very simple test. It's a 25x25mm square with rounded corners printed with three different settings for the "Shell thickness" (0.8 - 6.0 - 13.0), infill set to 100% and 35mm/s print speed. It should be quite obvious from these images what my problem is: Just in case someone thinks that rotation matters (other than for the reflection that would be equal): Can we agree that the first does not look like it has 100% infill? Please?

Sure. Here are my Cura settings http://flashgamer.com/a/infill/colorfabb_xt2.ini and here's a couple more images. As you can see from this one - the walls are quite transparent. If something like this label is pressed against it, you can read it quite easily. At the bottom, it's much harder to read since there is not enough infill - creating reflections. When held up against the light, it's more easy to see the difference. The infill refracts more, due to the air gaps. I'll switch to a yellow translucent PLA now and see how that works.

Hmmm.... That is very odd then. How could I possibly get the results I'm getting? I mean - that should use the same amount of material for both outlines and infill. How come there is such a visually noticeable difference AND only on the infill? J

Hi Daid, That picture looks nice, but it isn't showing any infill at all? From what I can see, this picture shows a normal print with a normally filled in top? The top and bottom fills just fine - also in my picture. In fact - since the top/bottom fills just fine, you cannot really see it in my picture - it's 99% transparent :-) It's only the infill (i.e. not the top/bottom x layers) that is lacking plastic. Not the outlines or top/bottom. Do you have some Colorfabb XT or any other translucent plastic in the office? You will need to use transparent plastic to actually be able to view the issue. Using transparent plastic, you'll very easily see where there is not enough infill.

Well, I'm sure you know Daid :-) It is nice to view the problem in Layer-view, but I still have the same problem, so how can I increase the amount of material on the infill? (As in going beyond what is said to be 100%) I can post an image of a Kisslicer-sliced print if you'd like to see the difference for yourself? It looks a lot better when it comes to the infill, but not so much in other regards. Cura is much better than KS, but for transparent parts, this is a major drawback.

Hi guys, I'm am not under-extruding. My infill speed is set to 0 and should thus be the same as other infill. There is no visible difference (on the bottom infill of the model) if I'm using 25mm/sec, 35mm/sec or 50 mm/sec. I actually did http://flashgamer.com/arduino/comments/colorfabb-xt-tests. My temperature/speed/flow rates are all fine and I know my Ultimaker original can handle this just fine. After all - it has done more than 3500 hours of printing. As for flow adjustments - I thought it was obvious that I did this just to see how much material was missing. I guess not? ;-) I am 100% sure that Cura is under-calculating the infill, but you'll only notice this when printing with a transparent material. If I use Kisslicer or Slic3r the infill looks MUCH better, so this definitely has something to do with how Cura calculates the infill. 100% infill on both of these programs will produce fully filled in bottom with no visible gaps of air like Cura generates.

Hi, Cura is fantastic, but I have a question/bug for you. 100% infill is quite far away from 100%. Have a look at this picture: This is a print of my Raaco storage compartments that has been set to a Shell thickness of 4.8mm and 100% infill. It's printed with ColorFabb's XT material that looks almost transparent when fully filled in. Looking at the picture, you can see that 4.8mm in along the edges, it's quite transparent. When it comes to the part that is filled in (along the bottom) it is VERY clear that 100% is really more like 60% or something? If I crank up the flow rate to 170% it starts to look somewhere near correctly filled, but that ruins the dimensions as it's depositing way too much plastic along the outlines. So my question is - isn't this a bug? If it's like this by design, I would very much appreciate the removal of the 100% infill limit so I can crank it up to 170% ;-)

How about doing like the Chinese? You ship as normal mail, but require that it's signed for at the post office? I think this is called Registered mail or something similar. Must be cheaper than DHL? J

Hi, Great to finally see something new in terms of support material, but I don't think this is going to work very well. The primary reason is that these are just a bunch of lines. If they fall (and they will!) you're entire print is ruined. It also causes massive amounts of retraction that adds to the total print time. In good old Skeinforge, there was a similar support material, but it was clever enough to connect the lines at the ends. Not every end, but every second, so that the extruder did a single zig-zag motion to create the support. This enabled you to keep a higher speed while printing the support. I'd very much like this feature changed to implement this support in the same way Skeinforge did. Being able to set how narrow/sparse to make the support would also be nice.

Hi, Just now I wanted to order two spare nozzles for my Ultimaker. The price would be 47.31 EUR. It's bad enough that the nozzles themselves are $15 each, but shipping two tiny bits of copper using DHL is just crazy. I really want to have the "Untracked mail" option back... I suggested this a few months back, but the reply was that this disappeared with the new Shop and that it might be added back. Can you? Please? Pretty please with sugar on top? :-P

Any news on this? It feels so painful to buy a €15 spare part and then pay €17 in freight... J

The simplest test is of course to see if you can push the filament by hand. You should never need to apply any excessive force for the filament to go through the extruder. It could also be something as simple as an incorrect diameter on the filament? If it's too thick, it'll get stuck in the Bowden tube with no filament getting through. This is common with cheap filament, but I've also had it happen to high quality filament too. Are you sure this is not the case? I've had my UM for quite some time and this one bit my tail a few weeks ago... J

After tons of discussions here on the forum and the on the Google Group, Ultimaking LTD added a cheaper, uninsured shipping option to the shop. Now I can't find it? As can be seen from the image - there's absolutely no need to ship this with a courier... :-/

Erik says in the comments here that it'll be put online soon Online Streaming Services No idea why it should take time though as I think it's just a one button click within Ustream to republish a live event? J

I have waited for a 14 hour print today and I just came home to see "Err: MAXTEMP" on my Ulticontroller. So much wasted filament... :sad: This obviously means that I have now used (and broken) both the cables in yet another signal cable (the one from the Thermocouple). I do use my machine a lot and I also have a spare cable (in my unopened Dual Extrusion kit), but honestly - this weakness must be addressed more permanently? J

Just thought I'd update this thread with the resolution in case others have similar problems. I was certain that it was the electronics, but after UM sent me a PCB to test with, I could conclude that different electronics didn't matter. Earlier tonight I stumbled onto the solution when changing the hotend to a brand new one. Some months ago I experimented with conductive ABS: http://flashgamer.com/arduino/comments/conductive-abs-from-repraper This caused me to have to clean the nozzle maybe 20 times and during one of these "cleanings" I must have pulled on the Thermocouple cord rather than push it out through the aluminium block. This caused it to be slightly yanked out of it's metal casing. It still measured temperature fairly correct, but it reported it lower than it did when I now pushed it back nicely in it's metal housing. The other part of my problem was that material didn't flow properly. I found this to be due to the white teflon part (above the PEEK) being damaged. I've taken the hotend apart so many times, but I never pulled out and checked this piece since it's not where the blockages occur. I had a spare one and with that in place, the printer now works nicely. I hope this can help others with similar problems

Thanks for suggestions! Today I finally had time to do some thorough testing. I tested all connections (no sign of corrosion), did a quick reflowing of the TC-board, tried a different cable and it's still the same. Next, I found another temperature sensor based on a DS18B20. I taped both of these to my heated bed using Kapton, set 50 degrees and observed. Both my cooking thermometer, the DS18B20-based temperature sensor and the Thermistor beneath the HBP showed 50 degrees +/- 2 degrees. I then set 70 degrees and all three followed. Next, I taped the two temperature sensors tightly onto each side of the aluminium block on the print head,wrapped it in aluminium foil to prevent any heat loss to the sides and did the same experiment. When the thermocouple reports 70 degrees, both the other two sensors report 50-51. When I set it to 100 degrees, the other two sensors report 68 and 72 degrees. The UM set to 125 returns 81 + 85 on the other two. If I point my IR thermometer up onto the aluminium block through a hole in the aluminium, it also reports the same temperature as the other two devices. Between each measurement, I've let the temperature "set" for 5 minutes and the readings are very steady. I'm at a loss. Unless anybody has other good suggestions, I'll contact UM support tomorrow. Below is a picture of how I set this up for measuring. J PS: sorry about cross-posting to the Google group. I just want to get the printer up and do some long prints while the family is off on holidays :wink:

Hi, I've had my UM for about 1.5 years now. I'm very happy with it and I have used it a l lot. Lately I've been having problems with my hotend (after testing some conductive filament http://flashgamer.com/arduino/comments/conductive-abs-from-repraper) and I need some help. My problem is that it seems that the extruder never reaches the temperature that is reported on my Ulticontroller. When the Ulticontroller says 220, I can barely extrude PLA by hand and ABS is impossible. When it reads 230, I can extrude a little PLA, but still no ABS. When I crank the extruder up to 260 (max) PLA flows fine but it's not fluid, and ABS is still super slow. I've tried to measure the temperature on the aluminium block in two ways: - using an IR meter - using a cooking thermometer that goes up to 300 degrees (K Type Thermocouple) When the Ulticontroller reads 260, these two will both report 185 degrees. I know that IR meters are known to be inaccurate, but when the thermocouple gives the same result and I'm having a hard time extruding all types of plastic, I feel that this the UM must be reporting this completely wrong. My UM is upgraded with latest extruder + extruder drive and it's been running pretty much flawlessly for more than a year. Does anyone have suggestions as to where to start searching for errors? Do the UM thermocouple "fade" with age?

I don't agree with you @snowygrouch. The first three are all good (Dual extrusion, HBP and enclosure). UM are working on Dual and enclosure is patented. They won't do HBP since customers may electrocute themselves if using 220V and the external 12-24V power bricks that can deliver enough watts are really huge and expensive. I can see why they don't want to do it, but they could have offered it as a DIY-kit? I've had my UM for 1,5 years now and my pulleys have never slipped. As soon as Bertho's new feeder arrived, it's been near fail-free. I would also say that the UM has an "ABS proven hotend". One third of my prints are ABS and they all come out great on my home-built HBP - much better than on Makerbots and comparable printers that I have available to test with. Only exception is maybe the MakerGear M2. It's very near UM in both print quality and speed, but is much less hackable. Direct Drive, Bed height, filament sensor, tensioners? Nahh, but I have to say that this auto-levelling looks kick-ass: :smile: J

It's a classic problem when using PHPBB. It's so widely used that hackers easily can automate attacks once there is a glitch. However - there's tons of ways to solve this: - Keep the script (fanatically) updated at all times. Hackers react within hours. - Implement a way to instantly "spam and ban" messages by any trusted user (that has 50+ good posts maybe?) - Remove all the 10 (!) mentions of PHPBB in the HTML source (this is basically the same as announcing that you're hackable) - Change the default script locations for login and register There's heaps more, but that should be a good working list to start with

Why don't you print something like this? http://www.thingiverse.com/thing:16921 That would solve your problem, wouldn't it? (or did you buy hundreds of meters? :-) )

chrili_rakete: got an update?