Jump to content

chrisw

Member
  • Content Count

    161
  • Joined

  • Last visited

  • Days Won

    3

Posts posted by chrisw


  1. I'll try to use "change filament" for this.

    Currently I'm forced to sleep near the printer but I have print jobs of 5-7 days.

    (Now that I got nGen to finally print reliably.)

     

    This is a good reason to install TinkerGnome's firmware. During a Pause you can navigate to Advanced --> Move Material. That way you don't even need to touch the filament or feeder. You just rotate the little white dial.


  2. Pause at Height works overall, but in Cura 2.1 the Pause code is written at the end of a layer, instead of at the beginning of a layer as it did in 15.04.x. This probably shouldn't make any difference, however there's now a weird movement line after the Pause.

    So the printer pauses, then goes back to old spot on the previous layer, leaves a small blob, then goes on to a completely different place to start the next layer, and continues from there as normal.

    If it's possible to eliminate that jittery motion it would be awesome. Right now I'm fixing it manually by removing the rogue movement line of the gcode, but I know I'm going to end up messing up a print or two in the long run if I do this manually (that's how it usually works when I edit code anyway)

    Windows 10 here, if that makes any difference :O


  3. Phil,

    You might want to try out the Robert feeder as an upgrade before class starts. It improves several things: more reliable feeding, more adjustability for different filaments, and easier filament changes. That would be a very affordable way to deal with the almost-universal gripes about the UM2's stock feeder.


  4. Is it possible that changes in layer height and/or changes in infill amount at different stages of the print will be available in 2.2, @nallath ?  

    Interface suggestions:

    In the 2.1 GUI when changing Infill Density %  or Line Distance mm, it seems that either one directly affects the other, however the changed value of one doesn't show in the other.   ie:  If I change from 50% to 25% the Line distance increases,  but in the Line Distance box I still see an unchanged length.  I think it would help new users better understand infill variables if the changes showed up instantly in the numbers there.

    Another not-intuitive GUI-related item is the Speed category.   I typically use one speed for all sections of the print, so I didn't think it was necessary for me to enable the sub-category settings like: Infill Speed, Outer Wall Speed, Inner Wall Speed, Top/Bottom speed.  But then when I started printing the file I noticed big changes were happening to print speeds.  So then I went back to Cura 2.1, enabled the viewing of those settings, and realized they needed to be manually normalized. Maybe by default all those speeds could be the same, or all those fields should be visible by default?     Otherwise more people will get surprises while printing I think.


  5. The most headache-free resolution would be to abandon ABS from hither forth.  It's a PITA with toxic fumes, and it likes to warp.   What parts are you printng?   Are you free to choose other plastics?

    If you want to keep dealing with ABS, please post some pics of the good and bad areas of prints.  And realize sometimes you get a strain of problematic filaments that won't ever behave the way you desire.   Upsizing the nozzle helps sometimes with temperamental filaments.   Tried a .6 mm yet?

    Also, which feeder is installed on your UM2? Default, lRobertl's, Bondtech, or UM2+ upgrade maybe? How is your feeder behaving? Main reason I'm askng is your 107% flow setting. That shouldn't be needed unless it's an attempt to fix a failing feeder


  6. I totally forgot about the zortrax m200 and have been thinking about that one too now.

     

    The Zortrax was tempting for a moment when they had their "Buy 36 rolls of filament and get a free Zortrax" deal a few months back. That's the only time I would have considered their proprietary system. Open source is just such a relief when it comes to hardware replacements, modifications, adjustments, and filament choices. That's why Zortrax wasn't on my radar for very long.

    The BCN3D looks attractive. I'd love to have one in my house along with my two UM2s to try/compare. But I wouldn't let go of either UM2. Whenever I receive an order for items that need to be printed, I get right to it with both UM2s operating full speed.

    GR5 was right about PLA. It prints very pretty. When I need more strength I use PETG. PC can be a headache because it has warping and hygroscopic characteristics that PLA and PETG don''t have. Also PC doesn't like to stick to a standard bed easily. There is one version of PC so far that can be printed well at 250C: eSun's ePC. It's strong and the UM2 works well with it, but there are no colors available yet, and everything has to be just right to prevent warping.

    CF filaments in my experience have been stiff overall, but they have more difficulties with layer bonding and it makes them undesirable for my uses. An XTCF20 roll has been sitting in one of my bins for about 1 year now. It's about 3/4 full. I just wasn't impressed.

    If my production demands increased tomorrow and I needed a third printer, I'd buy another UM2. Both of mine have paid for themselves already, and any issues they may have are usually easily resolved, especially since Ultimaker has FBRC8 here in the USA with parts and service center ready fast.


  7. THX a lot! just cleand the head and resart the printing!

     

    I had a similar issue a few months back, and it wasn't the nozzle joint leaking for me. It was coming from the top of my Olsson block, where the metal isolator screws onto the Olsson block.

    The cause for my leak was a loose spacer above the PTFE coupler. If there's not enough pressure on that part (mine was a home made plastic cylinder, yours is the shiny metal cylinder in your pictures), then melted filament leaks and works its way through the isolator threads.

    • Like 2

  8. What speed are you printing the first layers? Corners seem to peel up when they get pulled by the next layer of plastic as the nozzle turns the corner, puling in a different direction. I'd try hair spray on the bed and make sure layers 1 and maybe 2 & 3 are slow enough to stick well. Like 25 mm/s


  9. Those that struggle with the idea of auto-slicing may want to try this experimental pause button plugin:

    https://github.com/fieldOfView/PauseBackendPlugin

    Disclaimer: it is a hack in more than one way.

     

    Thanks for providing that.  I've been working with this pause/play for auto-slicing, and I can't see much difference when it's enabled or disabled in 2.1, until I fill the plate with 30 duplicate small items.  At that point I'm unable to select an item and relocate it on the build plate if the auto-slicing is enabled.  However I can click the Pause button and rearrange those items easily, then hit Play to have them sliced.   *except for sometimes when I hit Play and I only get a "Preparing to Slice...." message that others have mentioned.

    And of course the layer view takes longer when Auto-slice is disabled, but that makes sense.  I'll turn on slicing before going to layer view.

    Once again Thanks for the plugin!


  10. Post processing is handled only when you save the code. Not after every slice ;)

     

    Long delayed reply here... sorry. I had work + I had to play with Cura 2.1 for a bit. What you're describing seems to be a change from 15.x to 2.1 that I hadn't recognized. Thanks for pointing that out. I'm on the fence about that change. I'm sure it saves time while setting up a print in Cura, however now I can't view the start/stop points for a pause/restart in the Layers view (unless there are some clues I'm not seeing, if so please point them out), or see the path the nozzle will take to/from its pause position. That was handy sometimes.


  11. So now you're saying that if I disable auto slice I will have less warm days to do BBQ? That's not helping ;)

     

    Disabling autoslicing would reduce our health costs.  Small tablets slow down during the slicing process, especially when I have 4 pauses scheduled during the print, maybe a tweak or two at different layer heights, and several objects present on the Cura build plate.   Every minor change I make to object placement causes a reslice   Blood pressure goes up because I have deadlines.  This affects all of our health costs :O Imagine picketers chanting "Go green and keep health clean!"

    • Like 2

  12. Anyways, how is the esun working out for you? You using the same settings as above?

     

    Congrats on finding an amazon deal! Settings are very similar for eSun's product. Here's what I've been doing lately:

    nozzle: .6 mm

    nozzle temp: 240-245 depending on color

    bed temp: 75

    layer height: .25

    speed: 35 mm/s

    retraction: 50 mm/s, 6.5 mm

    fan: off

    filament: 2.90 mm

    flow rate: 100%

    movement: 225 mm/s

    • Like 1

  13. Is there another bed calibration that you do with the + modification?.    Does the reset to factory conditions on the sd card over ride the reset firmware on Cura ?.

     

    That's the calibration I was talking about, using the wizard from the UM2 LCD panel.  I don't know about the differences of factory resets using SD card vs Cura however I'd run the calibration again after resetting just in case.

    I have 2 UM2's here but no + kits.  I'm just thinking of things I know of that could have caused the problem you're describing on one of my machines, after complete disassembly and reassembly.  One of those was my forgetting to do the bed calibration.  The reassembly took a while, and once it was all finished my brain decided it's ready to print, and I forgot that step.  

    Another thing I'd recheck would be the two rods that carry the hot end.  Make sure they're both firmly popped in place on the 4 plastic boxes.  If one of those were loose it could change the height of the nozzle by a few mm.

    • Like 1

  14. would it be better to reduce the current now to be on a safer footing???. ie

    M907 E1000 ; motor current

     

    That's a good question. I think it depends on how you're going to use your UM2.

    I like keeping the current set to the original 1200 because it seems like it should do more at 1200. But so far I have no evidence of shortcomings with it set to 1000. My extruder performed well when I turned down the power. It still got quite warm, but not enough to soften PLA any longer.

    The "small" heatsink that I posted earlier keeps the temperature down for high speed printing with the current set to 1200. But it still gets pretty warm. That aluminum heat sink came off an old motherboard. From that same bin of old parts I also pulled a copper heat sink with heat pipes. It may be overkill, but the motor never gets more than lukewarm with that beast installed. Here's a size comparison of the two:

    20160408_171906_resized.thumb.jpg.df62282dbbef63eae7480cbb0bd78371.jpg

    20160408_171906_resized.thumb.jpg.df62282dbbef63eae7480cbb0bd78371.jpg


  15. You can also decrease the current to 1000 mA if it still is running hot.

    /Martin

     

    Thanks for the info, Martin! I am using S3D, but I had already removed the 'increase extruder current' line from that startup script. Your idea of reducing the current to 1000 mA worked. My printer made it through the entire print with no problems.

    That was with the retract speed set to 45 mm/s and with 6 mm retracts. It's a real torture test: I calculated there are approx 15000 retracts in that gcode, spread over approx 16000 seconds. So almost 1 retract per second for 4.5 hours straight.

    The 1000 mA setting was a very quick fix. But I always want more, so I tried adding a heat sink to one of my extruder motors. It made it through the same gcode with the extruder current at 1200 mA, and the retract speed increased to 60 mm/s, So maybe a heatsink is only useful for extreme conditions.

    Bondtech extruders rock :)

    20160402_225505_resized_2.jpg.07b11c232d566237da31da84543f2f5c.jpg

    20160402_225505_resized_2.jpg.07b11c232d566237da31da84543f2f5c.jpg


  16. 10mm/s is very slow.  I'd try bumping those speeds to 35 infill and 25 perimeter, with the same temp setting.  At 10mm it could be slow enough to cause oozing... maybe the melted material is falling out of the nozzle faster than it's being pushed

     

    Also, printing temperature is adjusted to suit the printing speed. If I'm printing fast I want a higher temp, slower printing: lower temp. For that reason I don't think it's optimal to have such a big difference between the infill speed and perimeter speed, because your temperature of 240 is constant during the print, and it may be an ideal temp for the 25mm/s infill speed, but way too hot for the 10mm/s perimeter speed.


  17. My Bondtech feeders have been in use for almost a month now, and like I said earlier in this thread, I'm really glad I bought two. These things push like crazy.

    That said, nothing's perfect for every job, and today I think I found the Achilles heel of the current Bondtech QR model. This problem only seems to manifest when using lots of retracts in short time frame with filament that melts at low temps (PLA).

    The motor gets hot when it's retracting repeatedly. Directly attached to the motor is a metal gear box, which acts as a heat sink... and carries the heat further down the line to the final drive shaft. The drive shaft heats up the filament enough to melt PLA and clog the system.

    My first experience this morning, while printing a P51 Mustang root wing part from 3dlabprint.com was air printing after about 1 hour of real printing. The cause was this blob at the beginning of the bowden tube. All other clogs ever have been at the nozzle end for me, but this time the drive shaft turned my PLA into a blob and kept pushing til it plugged the tube.

    lotsaretracts1.thumb.jpg.34c985aca0a362e68c66b8c160123cae.jpg

    I cleared that problem, then tried the print again with print speed cut in half. That didn't help. Maybe the motor still got hot, and slower speed kept the filament in contact for a longer time with the drive shaft. This is what happened:

    lotsaretracts2.thumb.jpg.b01e6af538c4fe3fae04f5b879eb0976.jpg

    Now I'm bypassing a lot of the retracts and hoping the print finishes. I think the Bondtech QR rocks, but to make it more bombproof, I think the next version needs a vertical or horizontal separation between the motor and the final drive.

    This airplane root wing print is not an easy one btw, so it is quite a torture test for any feeder i think. The author certainly did a great modeling job, but didn't optimize it for easy printing :p

    lotsaretracts1.thumb.jpg.34c985aca0a362e68c66b8c160123cae.jpg

    lotsaretracts2.thumb.jpg.b01e6af538c4fe3fae04f5b879eb0976.jpg


  18. I think it's important that you can "feel" the pressure you put on the spacer.

     

    The good news: There's guaranteed to be 0 slop in that joint for the nozzle when the head moves back and forth :)

×
×
  • Create New...

Important Information

Welcome to the Ultimaker Community of 3D printing experts. Visit the following links to read more about our Terms of Use or our Privacy Policy. Thank you!