I used my printer for first time less than a week ago, did a few small prints, all ok.
Today i am doing another print (on normal, using luban) and as it was laying out the base layer and first few layers there was a sort of popping / clacking noise.
Base layer seems to be ok, its now building up the layers and the noise has gone away. All modules seems to be doing their thing.
It was hard for me to locate the sound as i am deaf in one ear, but it seemed to be coming from where the head was located over the build plate.
Could it have been it was touching / pushing on the the heated build plate making it pop?
Could it have been the build plan under the heated layer expanding? i remeber it making a single pop when i screwed it together?
What else could it be?
thanks, i will keep an eye on it. it was just like a rice krispy noise but a lot louder
it would pop and then like 30 seconds later do another - just one a time.
Interesting now the print has finished a few interesting things to report:
it seems like the build plate calibration has changed in last few days (no new firmware was applied)
the head has marked the build plate
it has left marks where it was printing base layer like i see on folks posts about it bed sticking too much
the small items stuck much more than previously
tl;dr i think the noise was the extruder not managing to extrude because the head was pressing against the build surface it seems that filament or scratches (or both) are now embedded / melted into the plate - i am starting to wonder if the build plate surface is not heat resistant to the hot end touching it…
oh i should add this is the first time i have printed with the android controller. All my other prints were from the PC using it to ‘play’ the file.
–edit–
ok i did a standard automatic calibration and printed using the PC - first layers aok no hitting the surface, after this will try again from the android controller and see if issue reoccurs. Maybe i missed the memo i should calibrate every day?
oh and this print is on high quality (the one that did the weird thing was on normal) - just logging here so I can capture variables.
@WilliamBosacker thanks for the help.
Yes this 1.12.1 - put that on the day I built the A350 before doing any print and before my very first initial calibration.
Either calibration points can vary that wildly in 4 days of non use, non movement. Or there is material difference between driving print from PC vs the android controller.
As I said my z offset was perfect, i did multiple perfect prints (my first ever 3D print had perfect adhesion of initial layer and came off leaving no residue) , 4 days later it did this. No change in filament, ambient temp, temp settings, STL model, location, no touching of build plate. Nothing. Just 4 days and this time I sent the file to the controller rather than pressing play button in Luban UI on PC.
Or it utterly wiped the calibration mesh (is that the right term?) or z offset from memory for some reason.
Thanks will keep eye out for that. I was extremely diligent in screw and bolt tightening using process of of tightening kitty corners screws and bolts and revisiting. (Just like say doing bolts on a head gasket).
Any idea what causes the z-offset changes if it isn’t something loose?
Good to know, never used the manual controls…. Yet!
I plan to do some A / B testing to eliminate variables that I changed between good and bad. Time to get my thermometer out and my thermal camera to see if the machine actually operates consistent temperatures too.
Beyond that I will just recalibrate and do z-offset between each day, if I can’t assume it will be consistent. No biggie.
anyway i can report the clacking noise was from the gear in the print head trying to push filament when it couldn’t - i think it was the noise of the gear / filament slipping against either each other or something like that. seems ok and no long term issues.
Did you do the calibration with everything COLD… and grab the calibration sheet so that it dragged significantly?
then ran a build - the extra temperature on the Head and the build plate is enough to close the gap…
Do your Calibration after warming the head to at least 100C and the bed to at least 50C and only grab the calibration sheet AND then back it off 0.05mm - you can always reduce the z offset during the initial Skirt or Rim - better than dragging on the bed…
@WilliamBosacker it can be overridden and kept on via console commands. I haven’t done it but others have. @DGMultimedia Also I’ll have to disagree with heating the nozzle. That’s a good way to make craters in your print surface if you accidentally screw up.
The thermal expansion of the nozzle from 20C to 100C, over an assumed 7mm nozzle height, will become 7.01mm tall. It’s negligible and for safety reasons should be left cold in my opinion.
I believe the G1029 A gcode is a blocking command, still, that will block execution of further commands until completion.
However, the touchscreen comms use an asynchronous channel that bypasses the normal gcode routines and is unaffected by blocking gcodes. So if you trigger the G1029 A via the console (so that the touchscreen is not locked in the calibration interface), you may still be able to set the bed temps from the touchscreen?
as you asked, nope this was calibrate at machine been on but not used for an hour, print, everything fine.
Come back 5 days later, turn it on, do stuff in luban etc for about an hour. Print, head hit surface.
Literally the only thing i can think of that might have changed is its possible the magnetic sheet might have got nudged a 1 single mm - but as the issue was in the center not the edges. Either way I now to a calibration and z-offset if i come back to the machine after a few days, the issue has not re-occurred.
oh I see, that’s pretty sad that the command won’t work. As I said, I never tried it, only heard about it from other people. If an update is required to make that command work, it would be quite unwise to not implement it as a standard method in the touchscreen at the same time and update.
i bought a steel rule to use its edge to see where i can see light under it. does indeed seem the center of my bed is maybe 0.1 o 0.2 mm higher than the some (but not all of the corners). do i need use octoprints visualization to see what calibration thinks or should it be apparent in the mesh data?
It only happened the one time that I uploaded the gcode to the controller and used touchscreen.
As such i don’t have video. Since then I use Luban directly connected via USB to send the gcode using the play button (not upload it).
As such it is unclear what caused the issue, i don’t believe i changed anything in terms of z-offset, calibration etc between the time it printed ok and the one time the head crashed into the platform - but i can’t rule it out, it was only my 3rd print.
Once my machine is back together and I have done the other tests you asked for via my support ticket I will try uploading gcode and seeing if the issue is repeatable or not.
did a print, turned machine off, went for dinner
came back, turned it on
sliced a different variant of the model in luban (no settings changes, just small change in dimensions)
sent via wifi to the touchscreen and started to print
and on print the head buried itself into the lower left corner of the bed causing massive scratch this time (no popping, like scratched through to silver metal this time) and dragging the bed off the magnetic sheet.
I didn’t change a single setting - just the model.
i recalibrated via touchscreen and it is now ok - this time i know i didn’t change offset, and i am using the same gcode directly via luban and its printing fine.
What firmware are you on?
Did you a automatic or manual calibration?
May you share and observe your M503 (in terminal)?
It would be good to do that between prints too, to be sure your grid isn´t lost by power cycle.
