I received the same disconnection error as the sled was approaching the area it disconnected previously, but not the same spot as it was cutting a different path.
I am running firmware 0.81
Not sure if it is a gcode problem or something else. (gcode attached)
When you reconnect does it say āAlarmā over on the right hand side like when it normally first boots up?
If thatās the case then I think that the esp32 could be crashing. Anna saw that happen in testing the other day. Iām still trying to track down exactly why it happened, but I think that it has to do with the encoder cables being lose. My guess is that as the machine is moving to that area it is pulling on one of the cables in a particular direction which is causing it to disconnect.
My first recommendation is to give them a spray with computer duster and try the hot glue trick.
This is quite interesting to me. I, too, have found that my failures on test runs seem to occur right around the same place in the run - not every time but 5 out of 6 times. It has me wondering if the problems we are seeing with disconnects and read errors could be related to complex gcode or math errors building up over the course of processing a file until they cause a failure. I have no way to determine that myself, though.
I think that its more likely to be something physical about that area like āWhen the bottom left encoder cable is pulled down and to the right it is more prone to disconnectā so when cutting in a space where it is pulled in that direction it happens more often.
I have been investigating it from a software perspective so I made a version of the firmware which does everything normally but just doesnāt actually move the motors (the whole rest of the software stack is the same) and Iāve run many many hours of files that way (even in different locations with different wifi networks) without seeing the issue. Nothing is 100% for sure until we fix it, but Iām leaning towards it being a physical issue and not a software issue.
The machine stops moving when it disconnects. When I reconnect it says that the yaml file is corrupt and I should restart the machine. I did not notice it saying āAlarmā, but also not saying it didnt.
I will give the hot glue trick a try. It would make sense that something physical is causing this.
Thatās an error code 152 - tracing that back to the FluidNC code it is ConfigurationInvalid
And that gets thrown in lots of places, but for a handful of reasons related to either a problem in the YAML (as it says), or that the ādefaultā YAML is in use, which can be either because there was a problem in the YAML, or because of the CPU throwing a Panic Reset.
After glueing the remaining cables in place I tried to move the sled in the area that it fails and got this error message while adjusting the Z axis. āMotor Current on top right axis exceeded threshold of 4000ā. I am assuming this is the cause of my error, but do not know what to do about it.
After glueing the remaining cables in place I tried to move the sled in the
area that it fails and got this error message while adjusting the Z axis.
āMotor Current on top right axis exceeded threshold of 4000ā. I am assuming
this is the cause of my error, but do not know what to do about it.
what is your frame size and where is it that you get the error?
that error is saying that the machine is pulling very hard to try and move as
commanded, it can be caused by a mechanical problem in the arm, or it could be
that you are enough out of the zone where the maslow can move freely that the
belts are just getting too tight (the fact that itās an āarea that it failsā in
makes me think that it could be the latter)
The frame is about 8ā X 10ā with the error happening about 1 foot up and 1 foot right of center. This should be well within the working area.
I have noticed that the upper right (one that errors) belt is very tight while lower left is loose. I have also noticed when extending the belts that the lower right one gets a little tangled for the first few rotations of the spool but works itself. Would it make sense that there is a mechanical issue with the lower right causing the upper right to pull harder than needed?
Anything specific I should look for when I take it apart? And I assume re calibrate when I am done?
Just brainstorming here, but what was your fitness number? Also, have you tried extending all the way out and checking to make sure the belts arenāt twisted or something further on the spool? Not sure what getting tangled and then working out means exactly, but maybe the first couple of rotations on the spool are twisted?
The frame is about 8ā X 10ā with the error happening about 1 foot up and 1 foot right of center. This should be well within the working area.
correct, that should be fine.
I have noticed that the upper right (one that errors) belt is very tight while
lower left is loose. I have also noticed when extending the belts that the
lower right one gets a little tangled for the first few rotations of the spool
but works itself. Would it make sense that there is a mechanical issue with
the lower right causing the upper right to pull harder than needed?
yes, if there were problems with estimating the frame dimensions (i.e.
calibration) that could cause it to tighten things incorrectly.
can you go into more detail about what you mean when you say the belt is tangled
for the first few rotations of the spool? that sounds bad.
Anything specific I should look for when I take it apart? And I assume re calibrate when I am done?
if you had a good, working calibration, you would not need to recalibrate, but
since you donāt, you should.
in terms of what to look for. make sure the belts are on the spools correctly
and if you are taking the arm apart, see about moving the motor away from the
idler so that it rotates easily.
When I pull on the belt to get it unspooling, slack will develop in the spool. The next 2 wraps in the spool will slide down then back up causing the closest section of belt will be behind the next to wraps around the spool. I can send a video this evening.
I was assuming the bottom left is slacked because the top right should be taking up the slack, based on the over current. Leading me to the bottom right which unspools different than the rest.
I was assuming the bottom left is slacked because the top right should be taking up the slack, based on the over current. Leading me to the bottom right which unspools different than the rest.
If the top right is overcurrent and the bottom left is slack, that means that
the top right is pulling against the top left and bottom right belts, not the
bottom left belt.
which means that the frame dimensions itās working with (defined by the
calibration) are incorrect
I went to get a video of the bottom right belt getting tangled as it unspooled, and of course it would not do it. I ran a fresh calibration and the x and y dimensions changed by about 4 mm each and finished with a fitness of 0.757. I am currently thinking the belt must have had a kink or something that I worked out since the frame dimensions changed after calibration?
I could not make the error happen while moving the sled around, so I am hopeful that was it. I will try the cut again another time. I will report back after the cut.
I did notice it seems a little more creaky and squeaky and the teeth on the arms are showing some where. Is this Normal?
