Posts made by Chriss

@Phaedrux Thanks for the confirmation... I have expected that but you know.....

@Hernicz said in Better filament Load and Unload handling:

There's no way to cancel and I have to wait until unload.g finishes before I'm able to select another filament. I'm also unable to put any script into unload.g, because as soon as unload.g finishes running I have no filament assigned.

I have no idea how the unload.g actually works but I had a similar problem with a trigger and I used M291 to interact with the user and you can simply store values in a global variable if you need it later.

Here how I did it with a trigger to head up my hotend, retract, bla bla and the operator can select a new temp for the hotend just because you can not flush the nozzle with ABS when your toolhead is at pla temp etc

;echo "Filament Change triggered"
M291 R"Tool" P"Change Filament?" K{"OK","Cancel"} S4
if (input == 0)
    G1 Y1 F20000
    ;echo "Heating up"
    M109 S190           ; Set temp and wait for it
    G1 E-86 F600        ; Retract filament
    echo "Change Filament"

    M291 R"Tool" P"Filament Changed?" K{"Yes","Cancel"} S4
    ; Yes
    if (input == 0)
        G1 E84 F600         ; Extrude filament till nozzle

        M291 R"Tool" P"Filament temp for nozzle purge" K{"210","230", "260"} S4
            if (input == 0)
                M109 S210           ; Set temp and wait for it
            if (input == 1)
                M109 S230           ; Set temp and wait for it
            if (input == 2)
                M109 S260           ; Set temp and wait for it


        G1 E60 F400     ; Extrude to clean the nozzle
        M400            ; Wait for the end of extrude
        G1 E-18 F800        ; retracht filament from meltzone (revo)
        M400
        M104 S0         ; Extruder temp to 0
        M568 P0 A0      ; Extruder heater off
        M106 P5 S0      ; Turn the LEDs off
        echo "Done"

    ; Cancel
    if (input == 1)
    M106 P5 S0      ; Turn the LEDs off

; No
if (input == 1)
    echo "Chancel chosen"
    M106 P5 S0          ; Turn the LEDs off

I found it very convenient to use whether I use the DWC or the display.

Cheers, Chriss

@Phaedrux Hmmmm.... I'm still not sure what I should expect. (And I do not do that for pause either to be honest). I never ganged in the pause.g from absolute to relative movement for example. My idea would be to move the toolead to the center front retract the filament, wait for confirmation and extrude a lot of filament to flush the nozzle. Retract a tiny bit and continue the print after a interaction with the operator.
So the move to the front must be a absolute move so I need to change that because reasons. Do I need to change that back before I resume the print than? Or does the FW takes care of that? I guess so because the head needs to go back to the saved position....

Cheers, Chriss

@oliof This is clear to me and what is clear from the docu of M600. But what is with the rest? Will the printer than in the "Pause" state? How will it resume regarding G91/90?

Cheers, Chriss

Hi *,

I have a question regarding the inter print filament change drill.
I realized that the SuperSlicer likes to add a "filament change" code to the generated g code when it detects a logo etc.
I had a quick look at it and it seams to me that it generates a M600 at the needed layer to change the filament. I understood M600 that it executes filament-change.g or pause.g if the first one does not exist.
So what can I expect here? The M600 executes the filament-change.g and goes in "pause" than? And I can continue the print than with a "resume" via the DWC or the display? Does the firmware resume than at the same location and with the same G91/G90 as it was?

Cheers, Chriss

@dc42 Many thanks for the hint. That looks exactly like the solution I was looking for!

Cheers, Chriss

Hi *,

I have the idea to extend my 6HC based printer with some features. One idea includes the need to execute a external shell script on the PI, do make some tasks happening. Is there any way to do that? something like "M??? -P"/path/to/shell_script.sh".?

If not: Can we make it a feature request?

Cheers, Chriss

@jay_s_uk It seems like I instantly felt in love with BtnCmd.

Cheers....

@jay_s_uk

That is a bit strange, I have found this in one of my older configs:

;; Bed Corner temp sensor (2=Orange, 3=Brown, 4=Green, 5=Yellow, 6=Purple 7=Black, )
; Configure Bed corner temp sensor as thermistor on pin temp2
M308 S5 P"0.temp1" Y"thermistor" A"Bed-Corner" T100000 B4138

So at least the colours of the sensors have/had that feature. I can remember that I created the list of colours in the past.

I guess that we all agree that we want to change the colours here, don't we?

"BtnCmd"?

Cheers, Chriss

Hi *,

I can remember that there is/was a possibility to specify the tool colour in the DWC somehow in the config but I can not find it anymore. Could somebody give me a hint please?
The docu from M563 or M308 did not help me.

Cheers, Chriss

@dc42
Let me point out one more thing here: The printer I see this problem does not have this problem since some weeks anymore. But all of the print which the printer performed are below 100mm in Z. I'm not sure whether that is time ore high related. I just wanted to let you know my observation.

@chrishamm said in duetpluginservice(-root) not starting....:

@Chriss said in duetpluginservice(-root) not starting....:

Is there a am64 repo for that? I would like to play with that in a VM.

No. You could build it from the sources for different architectures but since it requires an SPI link to work I guess that's rather pointless.

OK, so no VM than, I will grep a other PI than for testing that.

@Chriss said in duetpluginservice(-root) not starting....:

I'm confused that I'm the only person reporting that, I used the most recent raspberry image from the duet homepage and I did not altered anything.

If anything went wrong during start-up, the journal should tell you. No idea why they were empty on your machine, though. They should be enabled by default. What variant are you using, 32 or 64-bit, with or without GUI?

64 and without GUI.

Cheers, Chriss

@chrishamm Strange, most strange.... Is there a am64 repo for that? I would like to play with that in a VM.

Am I right when I assume that we need to have the process running twice, one as uid 0 and one as the dsf? I will create two supervisord script to manage the processes till I found the reason for that behavior via the systemd.

I'm confused that I'm the only person reporting that, I used the most recent raspberry image from the duet homepage and I did not altered anything.

Needless to say that I want my init system back.

Cheers, Chriss

Nobody with an idea?

@droftarts Yes, yopu are right, the limit page says that. But I can remember from the top of my head that the limits are very often mentioned command sections too. Maybe the lack of consistency confused me here. I will keep an eye on it and let you know if I come across again.

Cheers, Chriss

Hi *,

I play with my "Duet 3 MB6HC MB6HC 3.5.3, Duet 3 Expansion EXP3HC EXP3HC 3.5.3 " a bit and I found that in the docu about M308:

This code replaces M305 in RepRapFirmware 3. In earlier versions of RepRapFirmware, sensors only existed in combination with heaters, which necessitated the concept of a "virtual heater" to represent a sensor with no associated heater (e.g. MCU temperature sensor). RepRapFirmware 3 allows sensors to be defined independently of heaters, up to a max of 32 sensors. The association between heaters and sensors is defined using M950.

How is it possible to use "M308 S55" when the docu speaks about "max 32 sensors?

M308 S56 P"mcu-temp" Y"mcu-temp" A"mcu"
Error: M308: Sensor number out of range
ok
M308 S56 P"mcu-temp" Y"mcu-temp" A"mcu"
ok

Do I oversee something? Or the the docu not correct? Or should I read it as "max 32 sensors per heater" and the "total max sensors" is not documented in the section?
Maybe I do simply do not get it...

Edit: And I just saw that I can not find the max fans in M950 too. :? I think that it would be nice to have the allowed range at the "Parameters" section .

Cheers, Chriss

@chrishamm Oh, they are not disabled as you see, they are "inactive". I enabled them manually (just in case) and rebooted a couple of times. It seems that they do nothing at all.

It is very strange to be honest, I see no reason why they shout not start even if I start them with a "systemctl start ....". I have never ever seen a daemon on a UN*X like system with never come back after you started it. (Well, at least not if it is not totally malfunction)
And I would expect at least any kind of message on stdout or in the system log, but there is simply nothing.
I did not thing to the image, I downloaded the latest version, dd it to the SD card, booted the pi and did a "apt update ; apt upgrade -y", that's it. I came across the issue when I tried to start the "DuetPi Management Plugin".

root@blimy:~/SD-Card/sys# ls -al /etc/systemd/system/multi-user.target.wants/duet*
lrwxrwxrwx 1 root root 50 Sep 19 17:17 /etc/systemd/system/multi-user.target.wants/duetpluginservice-root.service -> /lib/systemd/system/duetpluginservice-root.service
lrwxrwxrwx 1 root root 45 Oct 15 15:38 /etc/systemd/system/multi-user.target.wants/duetpluginservice.service -> /lib/systemd/system/duetpluginservice.service
lrwxrwxrwx 1 root root 41 Sep 19 17:17 /etc/systemd/system/multi-user.target.wants/duetwebserver.service -> /lib/systemd/system/duetwebserver.service
root@blimy:~/SD-Card/sys#

Edit: Changing the shell of the user dfs to bash is the only thing I have changed in the meantime. I needed to do that to su to the user to see how the bin behaves when I start it manually to see if there is any error report.

Cheers, Chriss

Hi *,

I just reimaged my pi with the latest duet raspi image which is connected to a Duet 3 MB6HC (1.0 or 1.01) but the two services "duetpluginservice" and "duetpluginservice-root" are not starting. The prompt does not come back when I try to start it manually with "systemctl start duetpluginservice.service".

The services are "dead" after a reboot":

root@blimy:~# systemctl | grep duet
  duetcontrolserver.service                                                                         loaded active     running         Duet Control Server
  duetpluginservice-root.service                                                                    loaded inactive   dead      start Duet Plugin Service (root)
  duetpluginservice.service                                                                         loaded inactive   dead      start Duet Plugin Service
  duetwebserver.service                                                                             loaded active     running         Duet Web Server

It seams to work when I start it manually:

root@blimy:~# /opt/dsf/bin/DuetPluginService 
Duet Plugin Service v3.5.3
Written by Christian Hammacher for Duet3D
Licensed under the terms of the GNU Public License Version 3

[info] DuetPluginService.Program: Settings loaded
[info] DuetPluginService.Program: Connection established
[info] DuetPluginService.Program: Plugin DuetPiManagementPlugin loaded
[info] Plugin DuetPiManagementPlugin: Process has been started (pid 1472)

It looks OK to when I do the same when I do it as the dsf user (after allowing the login).

There is nothing in the logs unfortunately:

root@blimy:~#  journalctl -u  duetpluginservice-root.service
-- No entries --
root@blimy:~#

Any idea what's wrong here?

Cheers, Chriss

@dc42 Thanks... that makes sense that it is not working as expected than. And that explains the behaviour too. Let me get that together somehow.

@Phaedrux said in Variables not working as expeccted in M558?:

Can you share the results of M115 please?

#FIRMWARE_NAME: RepRapFirmware for Duet 3 Mini 5+ FIRMWARE_VERSION: 3.5.2 ELECTRONICS: Duet 3 Mini5plus WiFi FIRMWARE_DATE: 2024-06-11 17:14:16

Sorry, my bad. Let me see what David has pointed out.