Homeβ€Ί JGAurora A5 & A3Sβ€Ί Getting Started & Troubleshooting

Print trail shifts whilst printing

245

Comments

  • MattMatt Posts: 314🌟 Super Member 🌟
    edited February 2021
    Duplicated, delete this please.
    Post edited by Matt on
  • MattMatt Posts: 314🌟 Super Member 🌟
    Hi, the screwdrivers came today: tomorrow we'll try the job. Let you know.
  • MattMatt Posts: 314🌟 Super Member 🌟
    edited February 2021
    hi. This is the tester I have: what should I choose to measure the driver? I am testing with a less "sophisticated" tester: the result, with ac courent, is 0,54 for the y axis. Wat should I do? Turn it to cw until 0.6? Pointing the black to the processor's gnd, I get 0,02!
    Post edited by Matt on
  • michalko99michalko99 Posts: 153Member, 🌟 Super Member 🌟
    hi. all is on provided link. As a GND point use USB socket-best to avoid shorts. Go to 0.6 without worries

    https://jgaurorawiki.com/a5/stepper-drivers


    Thanked by 1Samuel Pinches
  • michalko99michalko99 Posts: 153Member, 🌟 Super Member 🌟
    edited February 2021
    do not measure current. Measure DC voltage or you will fry them again
    Thanked by 1Samuel Pinches
    Post edited by michalko99 on
  • MattMatt Posts: 314🌟 Super Member 🌟
    I'm seeing that the x driver got a resistor (I gues it is that) burned::see pic. I'll change with a new one.
  • michalko99michalko99 Posts: 153Member, 🌟 Super Member 🌟
    it is ceramic bypass capacitor. But replace it anyway with new driver
    Thanked by 2Matt Samuel Pinches
  • MattMatt Posts: 314🌟 Super Member 🌟
    is this selector ok? I'm a complete noob in this,
  • MattMatt Posts: 314🌟 Super Member 🌟
    Ok, I figured it out (it was continous courent). Now I have to flash again the firmware?
  • MattMatt Posts: 314🌟 Super Member 🌟
    edited February 2021
    Ok. Same as before. Tried the same print, which appeared slant again, beginning from the same position (from 1/3d on). A tensioning issue? Perhaps opening so many times the cover, did the y-belt to go nuts? Or the vref of the Y must be higher than the table's? Any ideas?
    Post edited by Matt on
  • michalko99michalko99 Posts: 153Member, 🌟 Super Member 🌟
    hi Matt. Did you managed that your multimeter must be set one more click to right? Did you set voltage right? Your multimeter cables shall be: Black on USB socket, RED on driver touching trimmer metal part. If you measured approximately 0.5-0.6 Volts, it should be OK.Β 

    you can try Home all axes and after homing is finished, push hotbed by hand. If it starts slipping with low force applied, belt might be too loose. There is tensioner on axis.Β 

    Also provide picture of printed part how it was printed-showing printer, so I can see orientation
    Thanked by 1Samuel Pinches
  • MattMatt Posts: 314🌟 Super Member 🌟
    edited February 2021
    Hi Michalko. Yes, I tried with DC, and I measured several things before to start (a 1,5v aaa battery gave correctly a 1,52 or so voltage). I tried to measure with touching either the usb metal (that on the mobo), the cover metal and a separate piece of iron (such as a pinch): the results was of about 0,6. Though, I see that for the z driver must be set at around 0,8, due probably it supports more weight.
    The result is not so different from the prime one which I posted (I do not have it handy anymore, I trashcanned it): I must confess that, after I printed it (and I printed also another kind of model, a clip, to ascertain that the problem wasn't only in the slab model: same kind of result), I formatted the usb card and I put in it a gcode of the slab made with cura 4.8.0: when printing, the "hiccup-thuds" was still heard and the printer was like it missed to print some lines, until the remainder ones went anyway shifting. I tried to print the slab in vertical, and, albeit the thuds was heard, they was less loud and the print went apparently better, but it missed the final 10-20 rows (apparently 'cause they was printed upon the preceeding ones).
    I'm trying right now the slipping thing.
    Update slipping: the bed looks like it has some kind of small playing going to and fro. Perhaps I have to href the Z AND tension the Y belt.
    Post edited by Matt on
  • michalko99michalko99 Posts: 153Member, 🌟 Super Member 🌟
    Z axis Vref is higher because one stepstick is powering both Z motors in parallel, so it needs to supply higher current. If you have play in Y direction, it might be loose belt. I had to tighten mine once and no problems since then
    Thanked by 2Matt Samuel Pinches
  • MattMatt Posts: 314🌟 Super Member 🌟
    Ok. Later I'll open it up again and I'll try this tuning.
  • MattMatt Posts: 314🌟 Super Member 🌟
    edited February 2021
    Ok, tuned the Y to 0,75 and the Z to 0,80: this is the result:


    Post edited by Matt on
  • michalko99michalko99 Posts: 153Member, 🌟 Super Member 🌟
    sooooo. Are you happy with result? Because you stopped it on first layer. Also I see, that you have direct drive on your printer. This leads to excessive load on X axis. This was not mentioned on beginning, so you might be forced to increase Vref also for X driver. try to print something wavy, so your moves will cause some stress on drivers. there you will see if everything is correct
    Thanked by 1Samuel Pinches
  • MattMatt Posts: 314🌟 Super Member 🌟
    edited February 2021
    Hi. I stopped to first layer 'cause it began to thud and shift since the first one, the same way as I saw in previous attempt. So I considered unnecessary to continue until the end, also in order to avoid waste of material and to record a huge video (I only pasted a very short footage where you may see the shift and ear the hiccups).
    I have modified the printer as a direct drive, which didn't gave to me any issues so far (the stepper's the one which feeded the original bowden method: I tried to mount a slim version as a direct, but failed, 'cause evidently it didn'tΒ  had enough power to push/pull the filament... Or it didn't was the correct slim for this printer/mobo/drivers/firmware). Though I don't know what was the original vref of the z-drive when I bought the printer. The z-drive is currently at 0.8: should I try to add more vref? No overheating issues in raising it?
    The y-carriage, albeit didn't seemed apparently working badly, looks the one whence the thud comes: I raised the vref to 0.75 (should I raise more?). Its belt looks firm like a bowstring.
    All of components I had so far in the printer worked until I had to change one of the z steppers 'cause it was jammed: after I did that, and after I tried to flash the firmware for that stepper, I burned mobo+drivers and I had to buy & install anothe L-type mobo plus new 4 a4988: after I flashed the whole, all went amok. Perhaps should I try to flash another time? Though, I still don't know which one the correct firmware is: the jg guy who "assisted" me so far is in vacation for new year and I still wait for his firmwares... The previous "assistant" sent to me 3 versions and was substituted by him, 'cause he was evidently unable to solve the issue... Without telling me anyway which version I had to flash nor if there was some special workabouts to correctly flash it! Thank goodness there is this site, and I hope that I will solve with it.


    Post edited by Matt on
  • michalko99michalko99 Posts: 153Member, 🌟 Super Member 🌟
    Ok. But from your short and shaky video I do not see any Y shifting. More likely X. Also you have some nozzle to glass distance issue.Β 

    At first. Download calibration cube from Thingiverse. Scale it to 200% and start printing. Cube will show where might be problem. Especially if you have some light color filament would be nice to analyze. There is no need to finish it, just half is enough.

    Z axis should be OK, if you are able to move gantry up/down without hiccups(via LCD menu after homing).Β 

    For firmware. Best is to use community firmware from here, because you have full tutorial. But first print the cube..
    Thanked by 1Samuel Pinches
  • MattMatt Posts: 314🌟 Super Member 🌟
    edited February 2021
    The nozzle2glass issue you see is perhaps the reflex of the underglass from a perspective: the nozzle hits correctly the glass surface.
    The z axis runs smooth, when lifted.
    The community firmware... I don't even know how to put it in.
    Post edited by Matt on
  • MattMatt Posts: 314🌟 Super Member 🌟
    Nothing. Tried to print it, but the rumor became louder and more frequent.
  • Samuel PinchesSamuel Pinches Posts: 2,997Administrator
    > The community firmware... I don't even know how to put it in.
    Dear Matt, where have you looked???
  • michalko99michalko99 Posts: 153Member, 🌟 Super Member 🌟
    Hi Matt. So will you post result of cube? X symbol shall be facing printer front(display)

    Community firmware is here:

    https://jgaurorawiki.com/a5/firmware

    It can be downloaded for free and there is complete tutorial how to make it work. It cannot be described better.. It also includes troubleshooting when your Arduine IDE is new and you do not have required libraries already downloaded. I was doing this to my printer a few years ago as a arduino noob.Β 

    It just needs to read it several times to understand things correctlly. .Then you will get fully debugged firmware with tuned features. The best of it is accurate temperature measurement and babystepping when you update also LCD controller
    Thanked by 1Samuel Pinches
  • MattMatt Posts: 314🌟 Super Member 🌟
    • Dear Matt, where have you looked???
    Hi. I know where it is, but we remained that you were offering to input into my printer: but I stopped it in order to see whether she worked anyway with the ones provided by jg (also in order to see what the hell they did sent to me...). But we remained that we'll do this anyway, once all will be excluded: it is also for curiosity.
    • Hi Matt. So will you post result of cube? X symbol shall be facing printer front(display)
    Hi. I will post it when I'll be back from work. Yesterday I stopped to print it 'cause the print began to worsen since the prime layers as for thud & shift, so I figured that there was no need to waste anymore time & filament to ascertain that the situation was worsening even after driver tuning. I opened the case and I saw that the belt was in perfect tension: no apparent lubricant was needed to the rails. Though, I would like to lubricate them anyway: perhaps it will not be the solution, but I would like to try this too for extra satisfaction.

  • MattMatt Posts: 314🌟 Super Member 🌟
    edited February 2021
    Tried another print right now: I stopped after a while, due it began with thuds/hics whilst transversing the usual portion of the model. When the bed proceeded to harbor towards the back of the printer, there was these noises too. Perhaps it is a issue of belt/cog (I repeat that the belt looks very bowstring-like), or of the driver which isn't vreffed good and so overheats. I even lubricated the sliders of the bed.
    Post edited by Matt on
  • michalko99michalko99 Posts: 153Member, 🌟 Super Member 🌟
    edited February 2021
    try this at first....
    enough to print first 8mm, even when layers keeps shifting. You will lose around 2.5m of worthless filament.

    Also if you set in cura acceleration and jerk by yourself, post print screen here

    Lubricating LM bearings is not that easy. You have to remove them from rails, because inside of bearing is protected by collar to avoid dust getting inside
    Post edited by michalko99 on
  • MattMatt Posts: 314🌟 Super Member 🌟
    Ok, I'll try the print
    I didn't enabled acc/jerk.
    I sprayed the lithium lube upon the tubes.
  • MattMatt Posts: 314🌟 Super Member 🌟
    edited February 2021
    It is impossible to print anything. After tracing the perimeter without noises, the printing becomes noisy (it's like when the popcorn pops inside the closed pipkin...) and the bed jags as usual, when coming to tracing the diagonal filling, starting from a 1/5 of the filling itself. When I stopped the print, the head homed correctly, whilst the bed, when pushed towards its home position by hand, gave out the popcorn and jagged. I tried immediatly a new print, and the noises began since the very tracing of the perimeter...
    Briefing what I learned from the hypotesis made here by all of us, practically the issue must be given by one of the following:
    1) Belt exeedingly tensioned or untensioned (so, mechanical problem: but it is not in line with the behavior of the second print, which may hint to some driver fault...)
    2) y driver too hot or malfuncioning or not goodly vreffed (the noises of the second print might prove it: currently is set at 0,75)
    3) original firmware not good/not fitting or not goodly inputed
    4) what else?
    Post edited by Matt on
  • michalko99michalko99 Posts: 153Member, 🌟 Super Member 🌟
    OK. So maybe try to take apart printer again, disconnect stepper from mainboard and try to move bed with hand. What makes sense to me are 3 reasons: 1. bearings are dead-these factory LM10UU on A5 are crap, 2. idler gear bearing is dead. WhenΒ  worn, it holds belt correclty, but when it is moving, its rattling and changing its position and thus tension.3. overtightened belt. It just have to be tight, but not too much.Β 

    I strongly recommend to disconnect stepper from board, as during fast manual movement it can generate pretty much of energy and damage mainboard
    Thanked by 1Matt
  • MattMatt Posts: 314🌟 Super Member 🌟
    edited February 2021
    Ok. I have a spare driver: ironically, it is the one which still worked from facility, and which was into the y socket! Fortunally I've bought also another stock of 8 from ali: they will arrive next month (damn chinese new year...), but I'll try to endure. Anyway, it is very ennoying that one must still tune them manually (which could be even be done, if only there weren't the case opening tantrum...).
    I will try the belt sequence, then the stepper one (again...).
    How much vref could be sustained by an a4988? And, any ideas on putting a fan upon them (where do I must plug it?)?

    Post edited by Matt on
  • michalko99michalko99 Posts: 153Member, 🌟 Super Member 🌟
    you should not increase Vref above recommended value. It must work within margin described in tutorial. If it is not enough, you have mechanical problem which must be solved at first.. with disconnected motor you can feel by hand if there is some friction during hand movement. Factory A5 bearings are real crap, but it could be also bearing inside belt idler. What yet makes me curious, is cable management inside printer. Can hotbed cables catch against something during movement? By using flashlight you can shine inside of printer and look for movement during hand motion
Sign In or Register to comment.