Same fix for g28 y, change sign in front of the endstop pin, along with the location because the switch is at the back of the bed.
But…
Despite the machine being specified as a corexy. homing Y also moves X to the middle of the bed or beyond.
And a g28 xy went on to Z moving it down forever.
Fixed that, but I’m now convinced I have a bad board, it homes all 3 axises, but moves Y to the front while homing X, them moves X all the way right while homing Y, then homes Z correctly w/o moving the z2 cable from the 2nd extruder socket. Those 2 motors are belted together in the bottom of the machine, so it is possible the 1st Z motor was doing it all because the extruder 2 wasn’t energized, it not even described in the printer,cfg, which should have locked that motor. To be determined, but since that is about the same as I got from Marlin.
I’m convinced. unless you have a different opinion, that I have a bad MKS_Robin_nano_V3.1 board. It is a cast iron bitch to change, but I’ll do it tomorrow,
It sounds like the board is fine and you need to edit your config to make it work with your printer. This will require you to spend some time reading the Klipper documentation and go through initial startup and tuning.
this .cfg is set for corexy, which I have not mentioned lately.
I get to page 3 and STEPPER_BUZZ, then the results agree with my bad
board or wrong pins conclusion. setting =stepper_x gets a semi random
diagonal motion in a trapezoidal pattern that lack consistency, while
stepper_y gets me mostly x motion. Not pure, and in both cases more than
1mm as that document states, 3mm maybe. stepper_z works but again is
more that the 1mm bounce, 3 or 4 I’d guess.
These motion errors pretty well match the results when Marlin is
installed. That in my mind means the V3.1 pins aren’t the same as V3.0,
or the board is fubar. If anyone can find me a valid pin listing for the
MKS_Robin_nano_v3.1 board, I’ll check them in printer.cfg. The one V3.1
pdf I’ve found that shows the pins by names, shows them in 5 or 6 point
type, in yellow ink on glossy photo paper and they had to know that
would make them impossible to read even with a very strong glass.
What I started with is not a twotrees config because all of those are
for the V1.2 or earlier boards with different pinouts, so I started with
an example nano_V3.1.cfg from two years ago. MKS is, IMO way too close
to their chest with valid info for the whole Robin Nano family of
boards. I am encouraged, and have a second board to swap in, but this is
a lot like what happened to the V1.2 board after I resoldered a broken
wire on the x home switch while fixing a freeze up. The X home switch is
on the carriage, and in the way when working on the hotend.
Thanks Jake, take care & stay warm and well where ever you are on this
ball of rock and water.
There is a difference between the V3.0 board this is based on, and the
V3.1 boards I have. MKS says so on their web pages But they don’t say
what the diffs are. Helpful? not!
That’s the only thing I was missing in my printer.cfg
The STEPPER_BUZZ STEPPER=stepper_x still moves in about a 1x3 trapezoid
at a 45 degree angle
The STEPPER_BUZZ STEPPER=stepper_y moves x mostly and only about 1 mm.
The STEPPER_BUZZ STEPPER=stepper_z moves up and down about 4 mm with
the motor plugged into the 6th motor socket or the 4th motor socket
which seems to be paralleled with the 3rd motor socket on this v3.1
board. The 6th socket is labeled for a second extruder which this
printer does not have.
And what do I turn on to generate a klippy.log?
A “sudo updatedb && locate klippy.log” on the bpi5 running klipper finds
nothing.
Its nearly 400k and I had to cp it to /home/me/src/klipper to get it
someplace my security minded sshfs setup could see. Is updatedb somehow
enjoined from seeing it?
did that by copy/paste, then activate the enable pins after it would do
anything, made no difference nothing works now, mcu problems.
cleared the log, new one attached.
I’d be surprised if the |enable_pin| required to be inverted (!).
The drivers used for linuxcnc all default to enabled if that pin is
disconnected, not driven by anything. I’m not sure about the TMC2209’s
so I tried both ways, but cannot now talk to the mcu ack the octoprint
terminal, It did report with the ! sign there, that it couldn’t read the
x uart IFCNT. now with the !sign remove, the printer is reported to be
shut down but the motors for the 3 axis are enabled, so I’ll change that
back.
Inverting |dir_pin| is possible depending on your stepper wiring. Use
this to make sure the steppers turn the right direction
I’d be surprised if the |enable_pin| required to be inverted (!).
Inverting |dir_pin| is possible depending on your stepper wiring. Use
this to make sure the steppers turn the right direction
With enables back to where they were, inverted, the motor are now semi
free, acting like the coils are shorted, moving freely with a viscous
feeling. Temps both to 40C works, no motor response now.
TMC stepper_x failed to init: Unable to read tmc uart 'stepper_x' register IFCNT
TMC stepper_y failed to init: Unable to read tmc uart 'stepper_y' register IFCNT
TMC stepper_z failed to init: Unable to read tmc uart 'stepper_z' register IFCNT
TMC extruder failed to init: Unable to read tmc uart 'extruder' register IFCNT
This means that Klipper was unable to communicate with a tmc2209 stepper motor driver. This often happens when power is applied to the micro-controller, but not to the stepper motors.
Make sure power is fully up
Double check the wiring.
Double check that your board is really jumpered to UART
What driver modules do you use? And in my original config there was a typo: Extruder pin for E0 needs to be PD9 not PD8
If you want the email to stop show me how to attach a file to a reply done in fire fox. This is the first time I’ve tangled with a discord server and that function seems to be missing, I do note an upload button above now, but in normal browsing of discord messages, I do NOT see my own posts, and that upload button does not then exist if I’m reading someone elses msg.
Basically you are correct. My personal choice is to include some default settings in the cfg nevertheless. Makes it easier not to forget something when playing around and then wondering why things stopped working.
I took that out with # in front of it, restarted everything except the bpi5 itself and octoprint. it acts ok on the restart, but any attempt to move a motor is dead, octoprint terminal says: octoterm.txt (347 Bytes)
no motor has moved since I put the tmc2209 stuff in printer,cfg
wiith /tmp/klippy.log rm’d and /etc/init.d/klipper restarted, this upload in the first 240 lines of the log, with the last 6 lines being repeated at 1 second inteerval per line since. I have now stopped klipper because the log is over 2 megabytes already. Perhaps this head of that file will show the cause of the failure:
but discord won’t let me upload it, name extension error, picky picky
had to rename it to a .log, it is the head 240 lines. the next 2 megabytes is 2 line repeats at 1 second intervals.
And I repeat, this is based on the generic V3.0 version of this card, but the cards are v3.1’s not 3.0’s.
pin usage has changed according to the makebase site. Somebody with the chops to get a reply should be asking them, I did not get a reply when I did. I am being abused by their server though.
