SOLVED Z-axis not advancing mid print. Very random. Smooshes the lines out the side

Discussion in 'Calibration, Help, and Troubleshooting' started by Colin Oestreich, Oct 4, 2017.

  1. Colin Oestreich

    Joined:
    Jun 30, 2016
    Messages:
    3
    Likes Received:
    0
    Well.. From the beginning (bigbox titan hybrid model) this happened randomly. I have yet to ever figure out why it's happening. When it prints right, it's glorious, when it does this, it's a totally ruined print.

    Here we go. Dear maker I hope someone can help.

    Ever since I have gotten this printer, every once in awhile it would glitch and the z axis would not advance during printing. It has gotten worse over time (maybe.. really hard to tell because it's 100% random). To the point I know that about 75% of the prints will have this problem. (see image below)

    Example: I can print a file, and after 15 minutes or so (depending on print size or number of separate objects in the job), I can usually see that it has already started to mess up, but not always. Stop it. Restart and then it prints perfect, no changes at all.

    Here’s what I don’t get. I can start a print, it messes up. Stop the print, auto home, restart the print, and it prints fine (or not – totally random if it will print ok or not).. Other times, it prints fine from the first try.

    I’m literally at my wits end trying to figure this out.

    Here are things / theories I have tried. All failed.

    1. Bed leveling. I don’t see how that could be possible since 1 full print is fine and the next is a total waste. And if a 9 inch tall print can go fine, then I don’t see how it could be bed leveling whatsoever.

    2. Memory. Being an IT guy, I thought, well maybe memory is screwed up. So power down, leave it off for a few minutes, power up and try the print again (after a bad print). Seemed to work a couple times then totally proved that wrong, so it was coincidence.

    3. SD card.. Seems like it has problems reading the SD card from time to time. I thought maybe it was seated improperly. Checked it out up down and sideways, even tried printing with it slightly backed from being seated all the way in, and again, seemed to work a couple times, then was proved wrong again.

    4. SD card writing from computer & ejection. Thought maybe it had to do with ejecting the SD card before removing it from the computer after saving the file. Nope.

    5. 3D print layout. Early on, I thought maybe how the print was laid out on the print bed or maybe how it saved or something was it. Again, proved wrong very early on as the exact same print file screws up and then might print fine.. Totally random.

    6. Power.. Here is the one I have not tried. A 3D printer friend of mine (not with a bigbox) says he had issues with power regulation and ended up putting an UPS unit between his power source and the printer, which filters and regulates the power and it has done him a world of good. Interesting concept and I have considered purchasing an UPS unit, but I would much rather get support to interject first.

    7. Software / firmware updates? Honestly I don’t even know if this is even happening for the bigbox units. Suffice it to say, no updates, if they even exist, have ever been done. Ya, I'm sure a couple people will squirm about this.

    8. Overheating???? No idea. If it sat powered down for 2 days or just finished another print, still has the randomness of a good print or a bad print.. sigh.

    So.. any help would be welcome. I just have nothing left. I have ruined print after print and I cannot keep going like this. There has got to be something simple causing this. When it doesn’t screw up, it prints like a dream and I love my bigbox unit. Please help. I have dug through support and can’t find anything about this exact issue.

    In the example below and I failed to mention. Once it screws up, it screws up throughout the entire print. If it doesn’t fail after a bit the entire print will go just fine. Again, random and makes no sense whatsoever.

    Presentation1.jpg
     
  2. Ephemeris

    Ephemeris Well-Known Member

    Joined:
    Nov 25, 2015
    Messages:
    296
    Likes Received:
    182
    After the problem occurs, is the bed still level from left to right? If not, it could be coupler slippage. Several folks have had that bite them.

    Do you still have the Rumba with A4988 stepper driver boards? Check the voltage on it (Vref). The original instructions had it set pretty high and overheating is not unknown. If they overheat, they ignore commands until they cool down, which could produce your problem. If it's set correctly, try lowering Vref by say 0.1 volts to reduce the heat load.

    Checking the wiring, reseating the cables etc. might be a good idea.
     
  3. Colin Oestreich

    Joined:
    Jun 30, 2016
    Messages:
    3
    Likes Received:
    0
    Ephemeris,

    Yep the bed is still level, when it happens, both sides just don't move so z-axis just stays as the printhead keeps doing it's thing, so it just mushes.

    "If they overheat, they ignore commands until they cool down"
    ^^^^^^^^^^HOLY MOLY^^^^^^^^^ that really sounds like it could be the problem because that's basically exactly what it's doing.. ignoring commands.. I will admit it seems odd though that one print it screws up all the way through and the next does fine with no cool down in between, but if it's teetering on the edge.. who knows.. ugg.. so aggravating.

    Ill do my best to check wiring and such as well.
     
  4. Old_Tafr

    Old_Tafr Well-Known Member

    Joined:
    Mar 7, 2016
    Messages:
    561
    Likes Received:
    75
    Fascinating.

    Assuming a few things, like the same slicer file one time gives a good print and then a bad print or vice versa, then look at what elements of the system drive the "Z" axis.

    Also assumed that it's not mechanical, i.e. the coupling between the stepper motor and threaded rod is not slipping, otherwise the bed would soon be not level and accompanied by loud graunching noises as the bed twisted. This like anything mechanical would mean something had to be realigned/readjusted/leveled before starting to print again.

    Same argument for a stepper motor missing steps, as if the bed is level then they would both have to be slipping the same amount, this is most unlikely.

    Same argument for a bad connection between the Rumba and one stepper motor or the other. i.e. one would turn differently than the other resulting in a mechanical twist of the bed.

    So assuming the messed up part is the bed not moving down but still staying flat and level. The common factor here is that both Z motors are driven from the same stepper driver on the Rumba.

    Although variations in voltage and frequency and noise are not good, for this to only affect the "Z" axis would be odd, although this could not be eliminated entirely.


    Given that you have checked most things, my first stop would be checking the voltage setting of the stepper driver for the "Z" axis as below, step 27.

    I suspect the driver stops driving, it overheats or is set too low or the driver board is faulty. I'm not sure if the driver boards can be swapped, but there are five so at least one is spare, or swapping with the X or Y driver will move the fault around.

    (take anti-static precautions. (the driver board contain components particularly sensitive to static) and an acrylic BB is a static heaven) .......... preaching to the converted if you work on hardware in IT.

    I am not discounting a problem with firmware but I too have never updated it (and I'm going to install a Duet instead)

    Stepper Voltage

    https://wiki.e3d-online.com/wiki/Bi...y_Process#.2827.29_Tuning_The_Stepper_Drivers
     
    #4 Old_Tafr, Oct 4, 2017
    Last edited: Oct 4, 2017
  5. Colin Oestreich

    Joined:
    Jun 30, 2016
    Messages:
    3
    Likes Received:
    0
    Ephemeris and Old_Tafr.. I think we have the answer.

    We all agree that the z-axis not advancing from both motors and just stopping while the rest of the system keeps going (printing etc).. that there is some communication or issue with the z-axis control, whether it was wiring or the overheating issue causing it to "pause" until the driver boards cooled down or at least ducked below the threshold.

    So.. I checked the wiring. All is good (this was a factory assembly and they did a proper job!)

    Next, overheating. Well.. the main fan above the main board and driver boards.. hasn't worked properly since the beginning.. in fact, damn near all the fans have been shite from the beginning. a primary sore spot for me. So I removed that panel and the crappy fan. I put in a large fan (easily twice the size of the original) and plate I had purchased (with and independent power supply plug). I installed it where the old one was and it blows down onto the board and driver boards.

    After doing that.. not one single problem after multiple prints to test it out.. I think we have found the culprit.
    Many thanks for the replies and help and hope that this helps other folks out in the future!

    Soon as I get home I'll try to add a pic of the fan I added, but honestly it's pretty straight forward.. Just added a big honking fan to cool it!
     
    #5 Colin Oestreich, Oct 6, 2017
    Last edited: Oct 6, 2017
  6. Old_Tafr

    Old_Tafr Well-Known Member

    Joined:
    Mar 7, 2016
    Messages:
    561
    Likes Received:
    75
    Glad it is all working ! The killer in finding the solution is that both Z motors are driven from the same driver card.

    I did at one stage look into the driver chips used, alternatives etc. (Plenty of posts here on that) General info was that the type of chips used and the way they are mounted isn't exactly perfect in dissipating the heat. The solution is/was the fan in the lid to keep the driver boards cool, plus of course adjusting the driver card voltages correctly. There seem to be a whole "science" to setting the voltages, I simply set them to the voltage in the WiKi instructions.

    It's not good to find that fans failed, mine simply work, E3D work hard to get things right, but the BigBox was too successful, and I'm sure you have seen the story on their web pages.

    As above, to have easier control over firmware, a lot less noise without fitting dampers (especially needed apparently on the "Y" motor) etc. I have a Duet waiting to be fitted. I guess the Rumba was chosen to keep the cost down and it has its limitations.
     

Share This Page