Posts made by jannau

@barbarossa-cologne stall detection seems to be missing for the C motor, see E3D's config.g

a homec.g file is needed. It contains the instructions to drive the C axis until the stepper drivers detect a stall. E3D's sys files are on github, see their homec.g file

The Select Tool documentation doesn't reflect that. I'm just trying to get the documentation fixed. There is no issue with the actual behavior.

It is the following section

3. If all axes have been homed, run macro tpre#.g where # is the number of the new tool
4. Set the new tool to its operating temperatures specified by the S parameter in the most recent G10 command for that tool
5. If all axes have been homed, run macro tpost#.g where # is the number of the new tool. Typically this file would contain at least a M116 command to wait for its temperatures to stabilise.
6. Apply any X, Y, Z offset for the new tool specified by G10

Step 6. is since 2.03 already done in step 4., See the upgrade notes for 2.03 or dc42's comment in https://forum.duet3d.com/topic/16483/extrude-in-tpre/13

Hej,

https://duet3d.dozuki.com/Wiki/Gcode#Section_T_Select_Tool documents pre RepRapFirmware 2.03 behavior regarding when the tool offsets are applied. It says they apply only after tpost#.g is executed. It would be nice to clarify that beyond the firmware release notes.

Thanks

Ooznest has a modified DuetWebControl for the Workbee: https://learn.ooznest.co.uk/Guide/How+To+Use+WorkBee+Control/22

Source repository: https://github.com/lockryan/DuetWebControl/tree/workbeecontrol

@Danal said in RRF3RC1 - M585 probe does not stop when switch closes:

@jannau I take it you are running in CNC mode at the instant the probe occurs? I may try that as well.

yes, I'm using CNC mode. I just tested with aluminium foil that the probe move stops on contact.G31 K1 (in my case for probe 1) was helpful to debug the probe behavior.

I guess in your case the probe move is stopped by the machine limits or endstop and the tool offset is consistently set to 0.39 due to the Z trigger height of the probe. G31 K1 displays the Z trigger height.

G38.x might be another option for probing. It currently doesn't respect relative positioning. It is annoying for probing a work coordinate system but for probing tool offsets at known machine coordinates it should be less of a problem.

I'm currently building a E3D tool changer with a Duet 3. A couple of caveats:

• wiring kit requires recrimping and motor wires might be replaced
• mounting holes are incompatible
• sensorless homing will require tuning, I opted to add mechanical end stops for X/Y

I will only use 2 tools until extension board or the tool boards come available. You will need at least one external stepper driver if you want to use the Duet2. 5 internal steppers allow only 1 tool since 4 steppers are needed for motion and the tool changing lock.

M558 K1 P8 F60 C"!^io3.in"
G31 K1 P100 ; required for triggering with 3.0-b12
...
M585 Z50 F120 S1 P1
G10 P1 X0 Y0 Z0 ; probing is using T1, reset tool offsets to set WCS offsets
G10 P1 L20 Z0

Works here as as intended on a Duet 3 controlled CNC to set WCS.

There is M675 to probe the center of a cavity. This should simplify the probing for metal on metal contact. Unlike for CNC work coordinate systems there doesn't seem to way to set tool offsets relative to the current position. So without extensions it can't currently not be for setting tool offsets. If WCS are useable in 3D printing mode each tool could probe and use a separate WCS instead of tool offsets.

I opened issue 333 on github for the unexpected interpretation for the G38.x positions in RRF3.

G38.2 is only available in RRF3 beta11 and beta12. The Ooznest's macros are setting offsets for a work coordinate system. They use M585 for probing, reset the tool offset to X0Y0Z0 after probing and then set the offset for the work coordinate system. As far as I can see this is done because M585 is the only available probing tool in RRF2 for X and Y.

Have you seen the probing macros in Ooznest's Duet3d Firmware which are using M585 for workspace probing?

A single endstop/probe can be used for all 3 axes. M585 and G38.2 just probe a single axis at a time but the same contact is always triggered. G38.2 was just added to RepRapFirmware 3 and appears still a little buggy. I seems to accept only absolute target positions even after G91.

Fixed in duetwebcontrol 2.0.1

Hej,

I'm using a Duet 3 v0.6 + Raspberry Pi 4 as controllor for a CNC in CNC mode. I'm using the unstable apt repository and a recent update broke the display of the tool position in a work coordinate system. The tool position is not changed after setting work coordinate offset and seems to just reports the position in machine coordinate system. The machine position (after clicking on "Tool Position" in the dashboard) is always reported as "n/a".
This seems to be just a display issue. M114 reports expected work and machine positions and move commands are executed in the work coordinate systems. The regression started after following updates (extracted from /var/log/apt/history.log*).:

duetwebcontrol:armhf (2.0.0-5, 2.0.0-7), duettools:armhf (1.0.4.1, 1.1.0.2), reprapfirmware:armhf (1.0.4.1-1, 1.1.0.2-1), duetwebserver:armhf (1.1.1.0, 1.1.2.1), duetcontrolserver:armhf (1.0.4.1, 1.1.0.2), duetruntime:armhf (1.0.4.1, 1.1.0.2), duetsoftwareframework:armhf (1.0.4.1, 1.1.0.2), duetsd:armhf (1.0.3, 1.0.4)

Due to the multiple updated compoments I'm not quite sure it is a regression in duet web control. Is there a good guideline were to report bugs where it's not quite clear if they are in RepRapFirmware, DuetSoftwareFramework or Duet Web Control for Duet 3 + SBC setups?

1.0.4.0 is the current version of DuetSoftwareFramework, the software running on the SBC in the non-standalone mode.

The only option I currently see is start with a new Raspbian image and install Duet Software Framework as described here: https://duet3d.dozuki.com/Wiki/Duet_3_prototype_guide_for_OEMs

I'm currently doing that since my Duet3 apparently came without SD card and it seems to work. I have not connected any Hardware to the Duet yet.