Currents

Discussion in 'Motion System' started by orcinus, Aug 27, 2019.

  1. orcinus

    orcinus Well-Known Member

    Joined:
    Oct 20, 2015
    Messages:
    336
    Likes Received:
    113
    I thought the default currents are a bit high, even taking crazy accelerations into account, so i've tried reducing them a bit. Here's my setup:

    Code:
    M566 X400 Y400 Z8 C2 E400:400:400:400                                ; Set maximum instantaneous speed changes (mm/min)
    M203 X35000 Y35000 Z1200 C5000 E5000:5000:5000:5000    ; Set maximum speeds (mm/min)
    M201 X6000 Y6000 Z400 C500 E6000:6000:6000:6000            ; Set accelerations (mm/s^2)
    M906 X1400 Y1400 Z1330 C500 E1000:1000:1000:1000 I30    ; Set motor currents (mA) and motor idle factor in percent
    
    Noticing no issues or skipped steps so far.
    Ignore the crazy extruder acceleration and jerk - i'm trying to find the point where linear advance doesn't impede speed too much, and X/Y acceleration and jerk were my starting points.

    Now, one caveat.
    This will mess up your sensorless homing.
    There are two ways out - either change the currents to defaults before homing, then revert back after homing; or, spend an evening playing with the stall detection until you get it to work again.

    With these currents, X and Y are warm (instead of hot, as they were before), while the extruder steppers are just lukewarm. Z was always cool, so i didn't do anything there.
     
    Spoon Unit likes this.
  2. Spoon Unit

    Spoon Unit Well-Known Member

    Joined:
    Sep 6, 2015
    Messages:
    1,561
    Likes Received:
    483
    This is where I currently am with M566

    M566 X800 Y800 Z45 C2 E800:800:800:800 ; Set maximum instantaneous speed changes (mm/min)
     
    orcinus likes this.
  3. orcinus

    orcinus Well-Known Member

    Joined:
    Oct 20, 2015
    Messages:
    336
    Likes Received:
    113
    Thanks, will try yours.

    I was somewhat conservative with it until i get pressure advance tuned in.
    Which is for some reason proving difficult, because all the test prints show no difference even when swinging the pressure advance from 0 to 2s (yes, 2, not 0.2).
     
  4. Spoon Unit

    Spoon Unit Well-Known Member

    Joined:
    Sep 6, 2015
    Messages:
    1,561
    Likes Received:
    483
    Yes. I'm far far away from figuring it out.
     
  5. orcinus

    orcinus Well-Known Member

    Joined:
    Oct 20, 2015
    Messages:
    336
    Likes Received:
    113
    It's weird, it's as if sometimes it just outright ignores the M572 commands.
    I finally got it to work for one of the tools, the one with the 0.6mm nozzle, and the measured value is 0.92s, which is crazy high.

    Now trying to test it for tool0 (0.3mm) but keep getting same patches again.
     
  6. Spoon Unit

    Spoon Unit Well-Known Member

    Joined:
    Sep 6, 2015
    Messages:
    1,561
    Likes Received:
    483
    How are you measuring?
     
  7. orcinus

    orcinus Well-Known Member

    Joined:
    Oct 20, 2015
    Messages:
    336
    Likes Received:
    113
    Spoon Unit likes this.
  8. orcinus

    orcinus Well-Known Member

    Joined:
    Oct 20, 2015
    Messages:
    336
    Likes Received:
    113
    I don't get it.

    I've had the script output a gcode file that varies the pressure advance constant from 0.5 all the way to 3.0.
    I've double and triple checked the gcode, and it does what it's supposed to:

    Code:
    M572 D0 S0.500
    ...
    M572 D0 S2.975
    Yet the print shows absolutely NO difference throughout all of the layers.
    Is pressure advance constant somehow broken on toolchanging machines?
     
    #8 orcinus, Aug 27, 2019
    Last edited: Aug 27, 2019
  9. orcinus

    orcinus Well-Known Member

    Joined:
    Oct 20, 2015
    Messages:
    336
    Likes Received:
    113
    What. The. Hell.
    I just did a test print with pressure advance going from 0 to 9.
    The optimal constant is, apparently, between 6 and 7.
    SIX. And. SEVEN.

    No, not 0.6-0.7.
    6-7.
     
  10. orcinus

    orcinus Well-Known Member

    Joined:
    Oct 20, 2015
    Messages:
    336
    Likes Received:
    113
    That's on the tool that was giving me all those uneven extrusion problems, by the way.
    Any clues? Ideas? Could the bowden tube be pinched or something?

    I've checked the Titan and there's no backlash, and it spins freely.

    Might strip it and reassemble it just in case, though.
     
  11. orcinus

    orcinus Well-Known Member

    Joined:
    Oct 20, 2015
    Messages:
    336
    Likes Received:
    113
    That 7 figure is bogus.
    It's because the linear advance creates so much slowdown there's barely any difference between the low and high speeds - not because linear advance ends up evening out the extrusion, but because it ends up evening out the speed.

    Unless i'm misinterpreting something, it seems like the required linear advance corrections are so big that extruder isn't capable of delivering them without slowing down all the axes, even with the jerk and acceleration matching.
     
  12. orcinus

    orcinus Well-Known Member

    Joined:
    Oct 20, 2015
    Messages:
    336
    Likes Received:
    113
    What the...

    [​IMG]

    It's repeatable.
    On that tool.
    The height where it all goes wrong is different every time.
    I give up, i'm out of ideas for today.

    Edit: It would've kept me awake anyways, so i tore down the Titan on T0 and reassembled it. Absolutely nothing out of the ordinary. A bearing was not pressed down all the way into the lid, but i think that happened during disassembly, as i would've felt it anyways. Rotating the larger gear by hand feels smooth enough, but offers a slight bit more "graininess" on one half of the rotation. I doubt it's causing any issues, as i took it out of my BigBox where it was working perfectly (it's a hardened steel hob - there weren't any in the store when i was ordering the ToolChanger, so i cannibalized it).

    Assuming the stepper is fine, this leaves the hotend, i guess.
     
    #12 orcinus, Aug 28, 2019
    Last edited: Aug 28, 2019
  13. orcinus

    orcinus Well-Known Member

    Joined:
    Oct 20, 2015
    Messages:
    336
    Likes Received:
    113
    Aaaand suddenly it's fine again.
    [​IMG]

    Well, the extrusion is, i still have the same problem of linear advance seemingly having no effect.
    If this continues, i'm going to need something harder than beer.
     
  14. mhe

    mhe Well-Known Member

    Joined:
    Sep 5, 2015
    Messages:
    279
    Likes Received:
    77
    Out of curiosity, are you running vanilla PTFE or the Capricorn XS stuff? I put the latter onto my TC and it worked so nicely after only little tweaking that I haven't bothered dialling in the advance feature. 3mm retract, 0.3mm coasting, cooling fans on (after the config fix) and it spits close to flawless benchys all day.
     
  15. Spoon Unit

    Spoon Unit Well-Known Member

    Joined:
    Sep 6, 2015
    Messages:
    1,561
    Likes Received:
    483
    I'm using the vanilla PTFE at present. How does the Capricorn differ? Tighter tolerances ?
     
  16. mhe

    mhe Well-Known Member

    Joined:
    Sep 5, 2015
    Messages:
    279
    Likes Received:
    77
    1.9 inner diameter and less friction. Especially given how long our tubes are, it seems to make a difference, though I haven't tried with vanilla PTFE. I could continue to use almost all of my Bigbox settings, all I did was add 3mm retract and 0.3 coast and it prints almost flawlessly. And the TC is my first Bowden machine, so I thought I try to make it as easy on myself as possible and it worked quite nicely. When I come home from my work trip, I'll get the rest dialled in but I am already quite happy with the results.
     
  17. Spoon Unit

    Spoon Unit Well-Known Member

    Joined:
    Sep 6, 2015
    Messages:
    1,561
    Likes Received:
    483
    No stringing ? I am finding it tough to eradicate stringing. At present I'm on 4.75mm retraction and PA at 0.15 I think.
     
  18. orcinus

    orcinus Well-Known Member

    Joined:
    Oct 20, 2015
    Messages:
    336
    Likes Received:
    113
    I'm on vanilla PTFE.
    But! I've just noticed one thing.

    The little clip is almost falling out, the lock is out so far that you could fit two clips underneath.
    My guess is that that might be causing my issues - there's enough wiggle room for the PTFE tube to move up and down against the clip.

    And yes, it's in the hotend as far as it will go, and yes, i've cut it straight.
    I'll try to print another clip, or transfer one from a different tool, and see if two clips change things.
     
  19. mhe

    mhe Well-Known Member

    Joined:
    Sep 5, 2015
    Messages:
    279
    Likes Received:
    77
    My tubes don't move one bit. At the risk of stating the obvious, but I had struggled with the PTFE tubes on a direct V6 in the past, before the little clips were a thing. What helped me was keeping the tube pushed in and using pliers to lift up the black part as far as it would go. The clips that are standard now help just keeping it there. BUT - and that's a big one - if you don't keep the PTFE pushed in while raising the clip at the same time, you might create a gap where the tube is not in fully but the clip still fits.

    On older direct V6s it also helped to cut the PTFE tube to a conical shape so it would not create a gap between the end of the tube and the heatbreak. It should be fixed on all heatbreaks already for quite a long time, but if you maybe have some older heatbreaks laying around and used them in your build, then it might be a factor. The only hint these days that this problem ever existed is the revision history on the top right of the drawing: https://e3d-online.dozuki.com/Document/I2woGboCqjNEBkIx/V6.8-Heatbreak.pdf
    The "PTFE hole updated" thing fixed the issue which could be worked around by cutting the PTFE conical.
     
  20. Holten

    Holten Member

    Joined:
    Sep 9, 2019
    Messages:
    15
    Likes Received:
    6
    while I do have a couple printers I never really delved into customizing the configs (currents etc.) Is the main benefit here to reduce heat in the motors (aka reduce wear/tear)?

    I'm currently in the process of building my motion system/tool changer so I've been lurking in these forums for some tips.

    sorry off topic... @orcinus is that a Duck Jetfire keyboard I spy?
     

Share This Page