Major delta problems



  • I have a delta printer that's 1 meter tall with a 300mm diameter build plate. A few weeks ago, everything was printing fine. I obviously, however, managed to break something. I'm getting serious "ripples" in my print, but not the easy kind. They are "ripples" of what appear to be over and under extrusion (or probably better stated that my nozzle is drunk.) I think the best way to describe this is to simply show a print of a 270mm diameter circle, 0.2mm layer height, 1 layer. Print speed is 20mm/sec.

    The frame is aluminum extrusions (2040's) and aluminum corners. The extruder is a titan with E3D V6 hanging form a precision piezo orion. I'm printing simple PLA at 195C on a 65C plate.

    I get the same pattern on different build plates (glass with PEI on top, or glass with glue stick.) Something that's significant here is that the skin "fill" pattern is drawing lines from a 45 degree angle... There are two photos here, one showing the print on the build plate and one showing the print held up to a light (so it's easier to identify the areas of over/under extrusion.):

    0_1545704969746_20181224_210137 - Copy.jpg
    0_1545704981666_20181224_210606 - Copy.jpg



  • What makes this issue more interesting (to me, at least) is that if I print the exact same object, but configure my slicer (cura) to print the skin from an angle of 135 degrees (perpendicular to the above photos), then the "ripple" pattern is different:

    0_1545705229575_20181224_180120 - Copy.jpg
    0_1545705239453_20181224_180609 - Copy.jpg

    I've asked someone who's well regarded for his advice on this, but I suspect that he's never encountered something this fouled up. I was hoping that perhaps someone here has managed to mess up their own printer this badly and could share any solutions they had.

    If it helps, my auto-calibration results pretty good, with a deviation of only around 0.015 +/- 0.005.



  • I'm just throwing ideas in here and have no experience of deltas!

    I'd check over the heater cartridge and temperature sensor connectors to double check the heating isn't dropping out or spiking. Temp plot look normal? Another temp sensor available to test as a trial (while hoping it doesn't knock calibration out too far)?


  • administrators

    To me, it looks like over extrusion or possibly poor adhesion to the bed.



  • Some more information:

    This issue isn't limited to the first layer. In fact, it's more pronounced as the layer counts grow.

    The patterns appear random, but repeats across layers and prints. Any time I print a single layer disk with 45 degree lines, it'll look very similar to the one I posted earlier. The same happens for the 135 degree lines.



  • Gary, it wouldn't hurt to do an extruder calibration test. And the reason I asked about silicone lubricants on face book was to eliminate a bed adhesion issue.

    David and all, Gary's issues started with an odd mechanical issue (binding on one of the carriages) which he has mostly but completely resolved. I suggested printing 1 layer to see what effect that would have on the print as he was seeing smaller more localized top layer artifacts on larger parts. Rather than print a thick part, one can collect a lot of good data quickly on a single layer print. That's what led to this "discovery".

    The printer is an Ultibots D300VS+ modified to use the piezo rather than the stock FSRs. I wrote the assembly guide for this printer and it is an easily built machine that calibrates well. The bed is PEI and at 65°C, PLA should stick like glue (if it is not contaminated and clean).

    Cheers,
    Michael



  • @dc42 said in Major delta problems:

    To me, it looks like over extrusion or possibly poor adhesion to the bed.

    Checked the extrusion. 100mm for 100mm (+/- some fraction of a mm) over 3 tries. I changed filament, same issue. Bed adhesion is good based on the experience removing the prints. (PEI well cleaned with alcohol or acetone. I've also repeated the issue on glass with purple glue stick.)

    @doctrucker said in Major delta problems:

    I'd check over the heater cartridge and temperature sensor connectors to double check the heating isn't dropping out or spiking. Temp plot look normal? Another temp sensor available to test as a trial (while hoping it doesn't knock calibration out too far)?

    Temp plot is smooth.

    Something I think being missed is that this "pattern" is fairly consistent. If I print the same gcode file 3 times, all three prints have the same "ripples" in mostly the same area. (They might shift a mm or two, but it's consistent enough that I can recognize it quickly.) I'll admit I'm pretty new to delta printers, but I'd think that extrusion issues (if from the extruder or from the hot end) would be either a complete mess or more random.

    As well, and I'm not sure if it means anything, notice how the two prints posted above, differing only in the direction the lines are printed, look as if they are almost mirrors of each other. Not perfect mirrors, but I wonder if it's significant or not.



  • @garyd9 said in Major delta problems:

    nozzle

    Replace the nozzle. Seriously. On a Prusa Mk3 that had printed for about two weeks straight out of the box, printed VERY good prints such as the one on the left (and keep in mind that these are single wall prints, for flying radio control models; very challenging print) to the one on the right, suddenly and for no reason.

    I cleaned the nozzle a dozen ways, cold pulls, hot pulls, different filament, you name it. Still printed like crap.

    CHANGED the nozzle, and Viola! Printing just fine again. Put a $0.70 nozzle in it... 🙂

    Oh... and I don't have a pic... but one print in between really good and really bad just showed surface oddness, somewhat like yours.

    0_1545869225422_IMG_0104.jpeg



  • Something I think being missed is that this "pattern" is fairly consistent.

    Consistency makes me think the problem is with your motion-mechanics. If the problem was with the nozzle you would not be experiencing this consistency. I think that something new has either loosed up, or is now binding. Do you see anything wiggle when you attempt to move the effector around with one finger pushing under the hotend? Is your Orion mount still ok? Pulley grubscrews and effector rod connections still tight? Motor mount screws tight? Carriages still free? Also examine the drive belts carefully for damage or stretch marks.



  • @danal said in Major delta problems:

    I cleaned the nozzle a dozen ways, cold pulls, hot pulls, different filament, you name it. Still printed like crap.

    CHANGED the nozzle, and Viola! Printing just fine again. Put a $0.70 nozzle in it... 🙂

    Were the surface oddities consistent across prints? ie: same patterns?



  • @ayudtee said in Major delta problems:

    Consistency makes me think the problem is with your motion-mechanics. If the problem was with the nozzle you would not be experiencing this consistency. I think that something new has either loosed up, or is now binding. Do you see anything wiggle when you attempt to move the effector around with one finger pushing under the hotend? Is your Orion mount still ok? Pulley grubscrews and effector rod connections still tight? Motor mount screws tight? Carriages still free? Also examine the drive belts carefully for damage or stretch marks.

    I've completely replaced all three belts, completely removed all three carriages, removed the wheels from the carriages and reattached them. The carriages show good movement when attached. I've checked (and replaced) the pulley gears at the steppers, and the idler pulleys. I'm currently investigating if the issue might be related to the orion (but shoving so much teflon tape on to the orion->extruder mounting joint that I cracked my titan cover re-attaching it. I might also try to completely remove the orion from between the extruder and E3DV6 and see if that helps.

    Pardon me while I go order another titan extruder cover.



  • Pardon me while I go order another titan extruder cover.

    I replaced the clickety-clackety titan on my Kossel this summer with a Bondtech BMG. (After reading @deckingman's excellent review.) Do yourself a favor, you won't regret it.



  • @ayudtee said in Major delta problems:

    Pardon me while I go order another titan extruder cover.

    I replaced the clickety-clackety titan on my Kossel this summer with a Bondtech BMG. (After reading @deckingman's excellent review.) Do yourself a favor, you won't regret it.

    I plan on it, eventually. First I want to get all my existing problems solved before I go and change all the variables again. 😉



  • Tomorrow, I'll try replacing the stepper motor on the Y tower. The idea is that I was having good prints (even if the binding on Y was causing other issues) before I replaced that stepper, and as far as I've been able to recreate, the issue started shortly after replacing the stepper. It might be the motor, the wiring, or something I did while replacing the motor.

    It doesn't appear to be jerking or moving abnormally, but it can't hurt to try to replace it, and considering that it's a 0.9 degree stepper with microstepping set to 64, I doubt I'd be able to directly detect any flaw in the movement.



  • Replacing the stepper motor didn't help.

    Other things tried since my last post:

    I tried to rotate the build plate to see if the pattern would also rotate (which would indicate a build surface issue.) The result was... no change.

    I've tried to change the stepper motor current from 1 to 1.2 amps (and back to 1 amp): no change. (The motors are rated for 1.68.)

    I tried moving (in pairs) the ball cup arms, rotating them clockwise (so the arms on the X tower move to Z, the Z arms move to Y, and the Y arms move to X) - no change.

    Looking at how the pattern mirrors on the line of cartesian X=0 when printing 45 degree lines vs 135 degree lines, I closely inspected the carriage, pulley and idler of the Z tower. Nothing found, and no change after ensuring the idler was set up properly.

    Next steps: I might try rotating the effector, or I might try taking each of the three carriages off and moving them clockwise (the same way I did the ball cup arms.) If I can do something that moves or "rotates" the pattern, I think I'll be closer to isolating the issue.

    I might also try reducing microstepping from 64 to 32 (or even 16) to see what happens.



  • 65 is pretty hot (at least to me it is).
    PEI is awesome. Try dropping to 60 or 58.

    I got an MK3 recently. Love it. I try PETG recently on the MK3, Love it.

    I used it to re-print some parts on my dimond hotend mini kossel Duet Wifi.. Lovin it.
    I swear the more extruders, the more problems. lol


Log in to reply