Firmware bundle 3.3.0 released

chrishamm

We are very pleased to announce the next stable release v3.3.0.

This version brings many improvements like support for variables, accelerometers*, third-party plugins, and various bug fixes primarily for the Mini5+ and CAN expansion boards. In addition, many M-codes controlling system behavour are now supported on SBCs by the new DuetPi Management Plugin.

Please follow the upgrade notes before you install the new versions of

Users of Duets in standalone mode can upgrade using the ZIP file from the official RRF release. Users of Duets in SBC mode can upgrade using the normal apt update and apt upgrade process. The new packages are now available from the stable and unstable package feeds. Updated DuetPi builds for SBC users are now available, too.

Starting from this release, we will encourage the development of new third-party plugins for standalone and/or SBC mode. A first demo plugin allowing users to display the endstop states again is now available here. To install it, upload the entire ZIP file using the Upload&Start button or on the Files -> System page. After that, a new "Endstops" page is displayed on the Settings -> Machine-Specific Page. This installation process will remain the same for future plugins.

For interested devs there is now a preliminary documentation covering the basic plugin format used by DWC and DSF. .NET packages for third-party plugins on the SBC can be already found on NuGet and improved Python bindings will follow shortly on pip. Likewise, further documentation and guides will be added soon. For new third-party plugins we've also added a new forum category.

Besides, there is now a full documentation of the common object model available.

If you find this bundle is working well on your setup, please feel free to drop a comment below. If there are any problems or questions about this release, please create a new thread with [3.3.0] in the title.

Edit: New DuetPi builds are now available.

jMaddrell

Working well on my V-Core 3 with Toolboard 1.1 & SBC no issues so far. Is there a Swagger/OpenAPI spec file for the DuetAPI's? It'd make generating clients in other languages much simpler

war4peace

Installed 3.3.0 today, from 3.3 RC-3, everything went well, and simulating a file now updates the "Simulated Time" column.
I'm still to test the issue of long print jobs (40+ hours) not updating the "Print Time" column, but that's something that will only be visible after I finish the next long print job.

Touchthebitum

Hello
I'm trying to enable Plugin on the new 3.3.0 firmware.
sudo apt install apparmor worked
echo " lsm=apparmor" >> /boot/cmdline.txt give me an error message : -bash: /boot/cmdline.txt: permission denied

I can still activate plugin in DSW

Help please

tom33

@chrishamm I have the same problem since I upgrade with motor movement and 1XD (like in this post : https://forum.duet3d.com/topic/17780/clearpath-servos-with-1xd-expansion/76?_=1624015273855&fbclid=IwAR2bZCkl7tMWbKrGAJ7wN7WYqi2j4V8Fe9-WJHyBQuSh0yivL_SVDwbOHzs)

I make a gcode with just X movement and sometimes the motor change the direction...

dc42

@tom33 do you have a G4 S5 command early in your config.g file, earlier than the M569 command that refers to that 1XD board?

jay_s_uk

@touchthebitum the correct command is

sudo bash -c 'echo " lsm=apparmor" >> /boot/cmdline.txt'

Touchthebitum

@jay_s_uk thanks

tom33

@dc42 No I, don't have a G4 S5 in my config.g. I need to add this ?

dc42

@tom33 said in Firmware bundle 3.3.0 released:

@dc42 No I, don't have a G4 S5 in my config.g. I need to add this ?

Yes, that's often needed since RRF 3.2 to make sure that the tool boards and expansion boards have time to start up when power is applied, before the main board starts to configure them. Without it, the first command sent to a tool or expansion board (which is typically a M569 command to set the motor direction) may get ignored.

tom33

@dc42 I try this but still the same problem

chrishamm

@tom33 You need to put G4 S5 pretty much at the top of config.g. Is that the case? Also, have you updated the firmware of the 1XD to v3.3 as well?

tom33

@chrishamm I try this without success, the strange things is with the 3.2.2 it's ok, I also try with 3.3RC3+6 but no with the new one...

Some things looks weird with the pulse or DIR generator with 1XD

And the main board and expansion board have the the firmware

dc42

@tom33 you may need to adjust the step timings in the M569 command. Prior to 3.3, on the 1XD the timings you entered were all rounded up to the next multiple of 1.33us. In 3.3 the step pulse generation is faster and the pulse width entered in M569 is rounded up to the next multiple of 83ns. Additionally, the step to direction hold time is timed from the leading edge of the step pulse, so although the hold time from the trailing edge that you specify in M569 will always be honoured, the way that it is rounded up is different.

I will add a note about this to the upgrade notes for 3.3.

tom33

@dc42 I'll try this and keep you inform about the result.

I'm using leadshine CS-D508 Do you recommand any T parameters for this driver ? CS-D508_Manuel.pdf

dc42

@tom33 said in Firmware bundle 3.3.0 released:

@dc42 I'll try this and keep you inform about the result.

I'm using leadshine CS-D508 Do you recommand any T parameters for this driver ? CS-D508_Manuel.pdf

In theory, T2.5:2.5:5:0 should be sufficient to. But also try a little higher e.g. T3:3:6:1.

Tinchus

@chrishamm Whre can I find information about the use of accelerometers now in duet?

droftarts

@tinchus https://duet3d.dozuki.com/Wiki/Accelerometers

Ian

tom33

@dc42 I have just tried this values but same problem... there is no change between 3.3 RC+6 and 3.3.0 ?

dc42

@tom33 said in Firmware bundle 3.3.0 released:

@chrishamm I have the same problem since I upgrade with motor movement and 1XD (like in this post : https://forum.duet3d.com/topic/17780/clearpath-servos-with-1xd-expansion/76?_=1624015273855&fbclid=IwAR2bZCkl7tMWbKrGAJ7wN7WYqi2j4V8Fe9-WJHyBQuSh0yivL_SVDwbOHzs)

I make a gcode with just X movement and sometimes the motor change the direction...

Please start a new thread about this. In that thread post your config.g file and the results of running M122 and M122 B# (where # is the CAN address of the 1XD board) after you have provoked some incorrect movements.