3 weeks in, same issue



  • Hi.

    Yeah it's the issues with the blobs on the edges of the cube. I'm running a post process script that adds g10/g11 instead of the retraction In s3d. So all in all the firmware retraction is working.

    After I added m82 it stopped printing so that's not the issue.

    But I can not get rid of the blobs, Ive been at this for weeks and can't work this out. I've never had this issues with marlin based boards. I'm on the verge of giving up


  • administrators

    How long is the Bowden tube, and what hot end do you have?



  • I think your jerk/acceleration settings might be to high. Trying lowering them by ten percent and see if it improves.



  • @Plecto:

    Yeah it's the issues with the blobs on the edges of the cube.

    Might help if you post another photo or edit the ones you've linked and circle or otherwise highlight an example of "the blobs".

    @Plecto:

    But I can not get rid of the blobs, Ive been at this for weeks and can't work this out.

    Someone asked me "how's the 3d printing going?" the other day. I realized that any normal (sane) person would look at my current output and say "Wow! Cool, you printed that thing and it works and looks great! That's amazing!" Instead all I can see are the specific issues that I've become sensitized to and think, "It's complete crap, no point in printing anything until I can get this fixed…"

    I am famously bad at diagnosing printer problems but I ask that you highlight a specific 'blob' as I see several issues (solutions are just starting ideas, not the point of the list):

    • Regular vertical lines across the sides - stepper motor vibration? Need more steps/mm, loosen your belts, vibration isolate the steppers, sacrifice to the reprap gods? (This is my current problem and now I see it everywhere…)

    • Looks like a seam on the X-Y corner in the first photo so not sure what can be done about that, but at the top and bottom and both visible corners around the X in photo 2 the corners are 'flared out'. This matches the issue in the other forum post you linked to I think. Seems like an acceleration/deceleration issue as Qdeathstar suggests with the jerk/acceleration settings, tighten your belts :-)), adjust your speeds, dunno. The fact that the flaring looks reduced for most of the seam in photo 1 also points to ramping up/down speeds of filament and or print head in one or more dimensions.

    • Slight ringing as the head accelerates from the trailing (?) edges of the letters, slightly more on the 'X' photo. Could be same as point above, or the 'bouncy' artifacts (belt spring or similar) that this object is designed to emphasize.

    • Filament not laid down evenly layer upon layer, more so in the Y photo especially the bar 'underlining' the Y. Seems a lot less in the X photo, so something different in hardware or settings between those two axes?

    • Is that underextrusion for the layers at the center of the X face?

    • Filament flow does not seem entirely consistent within a layer - especially 2 layers above the top of the Y, or maybe it is just the lighting. Again less on the X view, but note the vertical variation on the right side corner of the X face. If it were my filament I'd blame humidity and cheap quality, but lots of your areas look fine.

    • Some more random blob artifacts at the 'V' of both the X and Y characters.

    • First layer definitely laid down different than the others :-))

    So there you have lots of different issues to nitpick and chase, and from your post I'm not sure which one it is (not that I have a solution anyway, just trying to clarify). Seems like there is a quality difference between X and Y axes, is there something accounting for speed, acceleration, jerk, belt, vibration differences between these?

    @Plecto:

    I've never had this issues with marlin based boards.

    If you have/had different results on same printer and just different board, can you map all the settings between the two firmwares? Some should be identical, some will have algorithmic differences that can be analysed.

    @Plecto:

    I'm on the verge of giving up.

    You could meditate on Ian's blog comment:

    @Deckingman:

    As a change from printing lots of test pieces with no practical value, I thought it was about time I made something useful.

    Sorry. :-)) But I did laugh the first time I read it. I realized last night that because DHL tried to deliver my new stepper motors at lunchtime on Tuesday, they missed the 'bag up' day and it will be another two weeks at least before I see them. But wifey loves the 'hook solution' I implemented for something the other day, and no one will ever look closely enough at them to see the vertical lines that annoy me.



  • @dc42:

    How long is the Bowden tube, and what hot end do you have?

    E3d V6
    500mm Capricorn Tube
    Bondtech BMG


  • administrators

    For 500mm Capricorn tubing I would expect 6mm to be sufficient retraction, assuming that you are using collet clips or similar to make sure that the Bowden tube does not slide relative to the collets when you retract filament.

    Do the blobs correspond to points where extrusion stops and a travel move starts, or not?



  • @dc42:

    For 500mm Capricorn tubing I would expect 6mm to be sufficient retraction, assuming that you are using collet clips or similar to make sure that the Bowden tube does not slide relative to the collets when you retract filament.

    Do the blobs correspond to points where extrusion stops and a travel move starts, or not?

    Ive been mostly using 5.7mm retraction and the results are mostly the same.
    The blobs correspont to points where the layer starts, so on end of travel move between layers.

    Ive been printing with outside-in direction, but switching changes nothing.



  • So it's blobs on layer change which mostly means that you just need to tune the retraction. Someone pointed out above that your jerk and accelerations are high. A quick squint at your config.g shows that for your extruder, this is the case. It's entirely possible that during retraction this could "break" or "stretch" the molten filament, rather than pull it it back from the nozzle tip. So try significantly lower extrusion acceleration and jerk and/or a slower extrusion speed. Also, you have a slight Z hop. Depending on how fast your bed moves, this could introduce a pause so try disabling that too.


  • administrators

    Do you have retract on layer change enabled in the slicer?



  • @dc42:

    Do you have retract on layer change enabled in the slicer?

    David,

    The OP stated above somewhere that he's using a post processing script to add G10/G11 commands in S3D (not sure why).

    Quick question regarding Z hop in G10 - does it do the retract, then move the bed or do they happen concurrently? If it's the latter case, then Z hop might slow the retraction speed down if the bed moves slowly?


  • administrators

    To put it another way: is the OP doing retraction on layer change, either by enabling it in the slicer and having the post processing script convert the retraction/reprime commands to G10/G11, or by inserting them in post processing?



  • Since this post process script cought your attention here you go.
    all it does it to change the retraction defined in s3d with g10/g11
    So ive enabled retraction in s3d set retraction distance til 1.00mm and 6000 speed.
    However these values do nothing since they are overwritten with g10/11

    {REPLACE "G1 E-1.0000 F6000\n" "G10\n"}
    {REPLACE "G1 E0.0000 F6000\n" "G11\n"}
    {REPLACE "G1 E1.0000 F6000\n" "G11\n"}
    {REPLACE "G1 E0.0000 F1800\n" "G11\n"}
    {REPLACE "G1 E1.0000 F1800\n" "G11\n"}

    and just to clarify ive used slic3r with firmware retraction and it produces the same issues.



  • Doing a test cube now with following settings:

    Force retraction between layers is uncheck in s3d

    M201 X900 Y900 Z60 E3000 ; Maximum accelerations (mm/s^2)
    M566 X600 Y600 Z60 E300 ; Maximum jerk in mm/min

    M572 D0 S0
    M207 S6 R0 F6000 T6000 Z2


  • administrators

    @Plecto:

    Force retraction between layers is uncheck in s3d

    So your retraction settings will have zero effect on those blobs. Try checking it.



  • Result:



  • Im printing another now

    Force retraction between layers is checked in s3d

    M201 X900 Y900 Z60 E3000 ; Maximum accelerations (mm/s^2)
    M566 X600 Y600 Z60 E300 ; Maximum jerk in mm/min

    M572 D0 S0
    M207 S6 R0 F6000 T6000 Z2



  • No changes. I've swapped filament. It seems like it's as good as it gets… 😕


  • administrators

    Are you printing outside perimeters first, or the other way round?

    Maybe it's time to enable pressure advance.



  • @dc42:

    Are you printing outside perimeters first, or the other way round?

    Maybe it's time to enable pressure advance.

    Inside first, however ive tried both to same result.
    Below example is inside-out

    M572 D0 S0.1
    M207 S4 R0 F4500 T4500 Z2



  • Apart from the obvious cooling issue, the 2mm z-hop seems a bit much, try changing that to 0.2mm (or don't use z-hop at all)

    M207 S4 R0 F4500 T4500 Z0.2 
    
    

    My experience:
    I used bowden setup at the time on my i3 clone (P3Steel), replaced it with a titan+pancake motor, the print quality difference was night and day 😉


Log in to reply