trying to understand diagnostics (M122)
-
hello all.
i'm pretty new to the duet ecosystem .
i was testing duet2wifi board on a bench and had many connection losses . so after searching and doing all the tricks to improve wif reception etc the issue remained .yesterday i loaned duet3 board from a friend , connected via ethernet as stand alone , and i had same connection loss issue .
then i started to realize that's something wrong here , and i found that the issue was with my bench power supply that was intermittently stop working and back on again for very short pulses .maybe it was overvolting also , don't know , already threw it in the trash and waiting for a new one .when i had those connection issues , i was expecting to see something in the diagnostics , but couldn't see anything .
for example , i checked under voltage events and its always 0 .
the board was sitting on a bench without printing anything , are those under voltage events logged only while printing ?
i understand that due to the power supply fault the board reset itself and m122 shows "fresh" data . then whats the use for this "under voltage events" then?another issue i had while testing steppers - i could "home" x axis without any endstop attached to the board .
it just gives me warnings "motor phase might be disconnected" and the axis was marked homed in DWC .
i could move the stepper without issues , even though it wasn't homed .
i wanted to test this on duet3 and it was the same - without any stepper motor attached i could "home" all axes , had motor phase disconnected warnings but the axes were shown homed .
i'm pretty puzzled why is this happening . i was also able to start a print (!) without homing any axis . this can be pretty dangerous for printers with powerful steppers and heavy gantries . -
@hackinistrator if the voltage was dropping so much that you got connection lost issues, I'm guessing the board kept resetting. Look for 'Last reset'. As far is the board knows, you turned it off!
Re axes marking as homed, all depends on what you have in your config for your endstops. If you have normally closed switches (which most do), they will trigger as homed if not connected. As far as the board knows, it was already sat on the endstop when you hit home! Far safer than normally open which will continue to move until you connect the endstops or hit the Estop!
Home-built CoreXY, Duet Wifi, Chimera direct drive, 2x BMG, 300x300x300 build volume
i3 clone with a bunch of mods -
https://duet3d.dozuki.com/Guide/Ender+3+Pro+and+Duet+Maestro+Guide+Part+3:+Commissioning/39
The firmware will do exactly what you tell it to. Best to commission it safely.
-
@engikeneer said in trying to understand diagnostics (M122):
@hackinistrator if the voltage was dropping so much that you got connection lost issues, I'm guessing the board kept resetting. Look for 'Last reset'. As far is the board knows, you turned it off!
Re axes marking as homed, all depends on what you have in your config for your endstops. If you have normally closed switches (which most do), they will trigger as homed if not connected. As far as the board knows, it was already sat on the endstop when you hit home! Far safer than normally open which will continue to move until you connect the endstops or hit the Estop!
what is the min/max pulse width for the under voltage event to register ? i would like to test that .
about the homing , i'm sorry , but this is not how its properly done .
of course i can write a better homing macro that will check endstop condition before .i think the default homing sequence should be better .
even my 6 years old 8bit board checks if the endstop is pressed before homing , and if it is , it tries to move the motor in the other direction to see if the endstop state changes . if not then something is wrong , and you hae an error.
duet board just marks the axis "homed" with some motor phase warnings .
is this behavior sounds normal to you ? -
@hackinistrator said in trying to understand diagnostics (M122):
what is the min/max pulse width for the under voltage event to register ? i would like to test that .
It's 1ms.
.i think the default homing sequence should be better .
even my 6 years old 8bit board checks if the endstop is pressed before homing , and if it is , it tries to move the motor in the other direction to see if the endstop state changes . if not then something is wrong , and you hae an error.
You can do anything you like in the homing files. We'll look at making the homing files generated by configtool do as you suggest.
