Hemera extrusion inconsistency

Discussion in 'HotEnds & Extruders' started by Travis Milum, May 11, 2021.

  1. Travis Milum

    Travis Milum Member

    Joined:
    May 11, 2021
    Messages:
    3
    Likes Received:
    0
    I've been trying to resolve this extrusion issue since receiving my Hemera kit. Individual layers are appearing with a very intermittent wave pattern.

    I'm using a Ender3V2 which has the 4.2.2 board. I've adjusted VRef for all stepper drivers. I have a dual-Z screw kit from E3D, BLTouch 3.1, Marlin 2.0 custom build with changes as specified by Hemera setup and mesh bed-leveling enabled.

    The pictures show 4 sides of a hollow 1mm thick, 20mm x 20mm wall calibration cube. This one was printed with Cura settings:
    Nozzle: 0.4mm
    Extrusion Width: 0.45
    Wall Line Count: 2
    Print Temp: 195
    Bed Temp: 60
    Flow: 100%
    Wall Print Speed: 25 mm/s
    Fan Speed: 75% (5015 radial with fang-style duct)
    Filament: BASF Ultrafuse PLA White

    Adjustments that I've made to test. None of which have had a significant effect on the waviness:
    Extrusion widths: 0.4 - 0.67
    Wall Line Count: 1-2
    Print Temp: 195, 200, 210
    Fan Speed: 0, 75, 100
    Wall Print Speed: 25, 30, 40
    Filaments: IID Max PLA+ Black, BASF Ultafuse PLA White, Filaments.ca Standard PLA 4043D Leaf Green, eSun PLA+ WHITE

    I've cleaned my rails and wheels, adjusted Z height, leveled X-gantry, all belts are tightened to low twang note, physically moving each axis slowly is smooth. There is no filament on extrusion gears or hobs. The extruder grip tension is unchanged since I received it since it seemed tight already. I've calibrated ESteps to 391.20 from the suggested 409.

    I've measured each of the filament diameters and they're all within 1.75 +-0.05 using mitsu calipers and micrometer. I've checked these tools for accuracy using metal calibration shims from mitsu as well. They're good to 0.01 and 0.002 respectively. Some filaments are on the low side and others are a bit higher, but nothing completely out of range. I've kept the filament diameter and density settings in Cura the same for all prints.

    The waviness is not the same on each face. It's not the same across prints on the same face. The wave lasts less than the width of any individual face, and wraps around corners. This is *not* the moiré pattern some threads discuss. I can actually see the moiré pattern behind these waves, but it is much less noticeable.

    I don't believe this has anything to do with issues on any individual axis. Z-axis binding would show banding on the similar layer heights. An X/Y-axis wobble would also show in the similar place on each corresponding traveling face.

    I'm printing on a borosilicate glass bed. It's surface is flat to within 0.01mm. See mesh visualization. This is using 3 tests per location, 7x7 grid.

    The only thing I can conclude is that there's an issue with extrusion, but I'm at a loss as to what would fix it.

    I know that my ender can do better than this. The stock extruder with a bowden tube produced some very nice benchies. Each with layer lines that were extruded perfectly consistent. I went to a Hemera to print flexibles, but if I can't do PLA as well I think I might have to try a different extruder.

    Has anyone else seen this? How do I make those waves go away?
     

    Attached Files:

  2. Travis Milum

    Travis Milum Member

    Joined:
    May 11, 2021
    Messages:
    3
    Likes Received:
    0
    Something occurred to me. Would voltage fluctuations cause this? Just did a test by heating the hot-end, and extruding a bit of filament and measuring the voltage across the red and black wires at the extruder. This looks like it fluctuates while extruding to control the unit. I've seen the values fly between +-7 volts and seems to typically read at +-0.25. Once the extrusion stops it'll stay at one position and maintain that voltage.

    The position was -3.35 volts. Then I turned on the bed. The voltage dipped by 0.02 then bounced to -3.38, eventually settling at around -3.36. Would a 0.02 volt change cause extrusion changes like what's in the pictures?

    I'll test again by turning off the bed mid-print to see what it does.
     
  3. Travis Milum

    Travis Milum Member

    Joined:
    May 11, 2021
    Messages:
    3
    Likes Received:
    0
    Turned off heated bed at layer 62. Turned off hot-end at layer 77. Turned hot-end back on at layer 81 (cooled to 175 pretty quickly!).

    There are about 3 perfect layers. I can't get a representative photo, but I think this is the issue. Once both the bed and hot-end are not causing power fluctuations, the extruder output is consistent.

    I should note that I measured across the wrong wires in the previous post. Measurements from A - A- and B - B- showed the same thing in any case.

    Need someone with more electrical experience to suggest how to address this. Add a capacitor somewhere? What would the rating need to be?
     
  4. Eviltoy

    Eviltoy Well-Known Member

    Joined:
    Sep 2, 2016
    Messages:
    73
    Likes Received:
    7
    Did you manage to sort this?
     
  5. Lombar

    Lombar Member

    Joined:
    Apr 29, 2020
    Messages:
    16
    Likes Received:
    1
    I have the same problem.
    Crazy for the problem I discovered a possible cause of this. If you watch the photos you must agree with me that there is a problem with the gears alignment. Photos 1-2-3.

    This causes that everything is full of debris and sometimes, with the normal dilatation and the esfort of move the filament, the gears in the tensor part are braking and make dents in the filament. Photos 4 and following.

    I opened the ticket 64421 about this and I hope Technical service stay working and not in hollydays. I try to lock for the part but I can't find it, and in a few photos I found the same problem. Looking them, the small gear in the same axle that the big gear, must be against the filament throat and not against the big gear, isn't it? Other way it's very difficult that work as somebody designed it.
     

    Attached Files:

Share This Page