Mesh levelling stopping about halfway through G29
My BLTouch seems to be working, homes fine, but it can't complete a G29 mesh. It stops with the warning g29
Error: Probe already triggered before probing move started.
If I immediately run G29 again, it starts fine but stops again usually at the same point about halfway through the mesh..
Any ideas?
35 Replies
Increase horizontal_move_z: by 1 or 2mm. For example in your printer.cfg add: To the user overrides section
extended-salmonOP•2y ago
I... don't understand what that means sorry..
Do you have a vcore3? With ratosv2
extended-salmonOP•2y ago
VCore 3 with Duet 3..
Ok well edit the printer.cfg file, like you did when you first set up the printer. Scroll down the file until you reach the user overrides section. Check to see if there is already a section under that that has [bed_mesh]. If it's not there add it. And under [bed_mesh] add the line: horizontal_move_z: 6 (like I show above) and save the printer.cfg file, and restart the firmware. That should fix your problem.
Duet is RRF not klipper - so this isn't going to work
Sorry!
In that case you need to alter the z height of the hotend during the bed mesh operation so that it is higher than it is currently. (During travel between mesh points)
extended-salmonOP•2y ago
The thing is, it probes fine up to a certain number of points, so I can't see how changing this will help..
Perhaps that point in the bed is high enough to not allow the bed probe to trigger properly.
extended-salmonOP•2y ago
Nope, it fails at different points randomly, some earlier some later..
I am currently trying running G29 S0.. Can you tell me what the S component is?
Sorry that's all I've got. Not familiar with RRF or the Duet so I should have kept quiet lol. Hopefully someone will help you with a similar setup
extended-salmonOP•2y ago
No worries, thanks for trying..
G29 Mesh Bed Compensation (RepRapFirmware)
RepRapFirmware:
S0 (default if no S parameter) Probe the bed, save the height map in a file on the SD card, and activate the height map. The default folder for the height map file is /sys and the default file name is heightmap.csv.
S1 Load the height map from file and activate bed compensation. The default folder and filename as for S0.
S2 Clear the currently-loaded bed height map
To define the grid, see M557.
In RepRapFirmware 3.2 and later, G29 runs macro file mesh.g if it exists, otherwise it behaves like G29 S0. The mesh.g file can perform other actions (e.g. homing or tool selection) and then use G29 S0 to do the probing.
maybe @oliof can help?
extended-salmonOP•2y ago
He's on it,I believe..
@billyd60 in RRF its all in G31 and M558
this looks, by and large correct, 2.7mm trigger height, 6mm dive height.
extended-salmonOP•2y ago
I'm beginning to suspect faulty probe or wiring..
@cairn4177 looking at Troubleshooting BLTouch (https://docs.duet3d.com/User_manual/Troubleshooting/BLTouch_troubleshooting) documentation, you may want to add 0.2sec settle time to the probe definition, so add
R0.2
to your probe definition so it looks like
M558 P9 C"io7.in" H6 F100 T12000 R0.2
(also see my comment about M501 in the #rrf-duet channel)extended-salmonOP•2y ago
Added the R0.2 but still no dice. Is it worth trying r0.4 ?
only if your motion system really does not settle quickly. dive speed is already slow (F100) so I don't think if 0.2 helps 0.4 will
anything in config-override.g?
extended-salmonOP•2y ago
Nope..
okay
then I would point at wiring/loom since it seems to fail at the same point, it may be something is dragging on the wire.
extended-salmonOP•2y ago
Yeah, going to take a look there.. Thanks! 🙂
Incidentally the Z Probe status in the Duet browser blinks pink 1000 very briefly on a few points before failing and remaining pink 1000. If this indicates anything in particular?
OK, so I had an intermittent open circuit on the white wire to the BLTouch, and eliminating that solved the G29 failure, but... if it ain't one thing it's another, and now my Y axis endstop doesn't light up when tested, and when I home all or home Y seperately Y thinks it's already homed. I can bluff the machine into running by having the Y axis at zero at power-up, and accepting the false Y-home result, but this is obviously not ideal. What's the best way to test an endstop that doesn't light up?
that also indicates bad wiring as they are normally closed
extended-salmonOP•2y ago
By 'open' I just meant I removed the wiring ribbon and tested each wire. The white wire had a break somewhere, and when I replaced it the G29 fault wasn't repeated.
yes
I was talking about the endstop now
you need to make sure the wiring is right since Duet changed the endstop pin order so you dont burn your input when you plug in the endstop flipped
extended-salmonOP•2y ago
But this has worked fine for a long time. Nothing has changed..
Well, the thng that changed is the Y endstop light doesn't blink when you test it..
that indicates some wiring issue or, rarely, a broken switch
for a quick test you could unmount the y endstop, connect it to X endstop wiring, and see if it works there. If yes, its the wiring. if no, its the endstop
extended-salmonOP•2y ago
Can I swap the X and Y wsitches to test?
see above
yes
extended-salmonOP•2y ago
😀
alright I need to grab 3 hours of sleep before flying back home
extended-salmonOP•2y ago
Thanks friend!
(disassembled and reassembled, works now. The dreaded 'intermittent fault'..
Today's Hassle: Heater 1 (hotend) not heating. Error: temperature rising too slowly.. must be wiring I guess..
no
you need to pid tune again I guess
unless its not rising at all
then its either the thermistor or heater