Duet2 Wifi+Duex5 and SBC

taconite

Yes the GND1 of the little PCB from @deadwood83 has a connection to the GND of the Duet and the SBC

So what I did:

Removed the Shield from one side
connected the GND1 pin of the little PCB with an extra cable to the GND of the SBC
reduced the spifrequency to 2 MHz.
--> same thing as before

What do you mean with "bit long in the tooth"

dc42

The Duet to Pi cable is way too long. Also, as it is a round cable, it will have higher capacitance and more crosstalk than a ribbon cable would. You should replace it by a short (150mm or less) ribbon cable.

taconite

Changed to a short ribbon cable - nothing changed - still "no header" fault message

taconite

maybe I should try the duetpi-lite?

taconite

@dc42 Sorry for bothering you again. Do you any other ideas?

wilriker

Sorry for not getting back to you earlier. DuetPi-lite won't make a difference.

Can you please run DuetControlServer from command-line with -l debug to use debug log level and post the output? Former output posted here did not contain enough information.

taconite

@wilriker said in Duet2 Wifi+Duex5 and SBC:

Sorry for not getting back to you earlier. DuetPi-lite won't make a difference.

Can you please run DuetControlServer from command-line with -l debug to use debug log level and post the output? Former output posted here did not contain enough information.

No problem at all! Don't feel sorry - every help is appreciated!

With what command line exacly? Sorry I am a bit of a noob with RPi and Shell

wilriker

@taconite What I meant was a command-line shell, i.e. login via SSH or in case you have keyboard and screen connected to the RPi open a shell there (in UI look for Terminal or Shell). Basically the location where you also ran systemctl status ... or journalctl.

taconite

@wilriker said in Duet2 Wifi+Duex5 and SBC:

@taconite What I meant was a command-line shell, i.e. login via SSH or in case you have keyboard and screen connected to the RPi open a shell there (in UI look for Terminal or Shell). Basically the location where you also ran systemctl status ... or journalctl.

I guess we had a misunderstanding. I understood you that the command line needs to be in SSH or on the pi command itself. I was wondering about the command that I should issue To phrase it differently I don't know which command I should run.

pi@duet3:~ $ systemctl status duetcontrolserver.service -l debug
Unit debug.service could not be found.
● duetcontrolserver.service - Duet Control Server
   Loaded: loaded (/lib/systemd/system/duetcontrolserver.service; enabled; vendo
   Active: activating (start) since Thu 2021-02-11 20:28:52 CET; 1s ago
 Main PID: 1596 (DuetControlServ)
    Tasks: 9 (limit: 2062)
   CGroup: /system.slice/duetcontrolserver.service
           └─1596 /opt/dsf/bin/DuetControlServer

Feb 11 20:28:52 duet3 systemd[1]: Starting Duet Control Server...
Feb 11 20:28:53 duet3 DuetControlServer[1596]: Duet Control Server v3.2.0
Feb 11 20:28:53 duet3 DuetControlServer[1596]: Written by Christian Hammacher fo
Feb 11 20:28:53 duet3 DuetControlServer[1596]: Licensed under the terms of the G
lines 1-13/13 (END)...skipping...
Unit debug.service could not be found.
● duetcontrolserver.service - Duet Control Server
   Loaded: loaded (/lib/systemd/system/duetcontrolserver.service; enabled; vendor preset: enabled)
   Active: activating (start) since Thu 2021-02-11 20:28:52 CET; 1s ago
 Main PID: 1596 (DuetControlServ)
    Tasks: 9 (limit: 2062)
   CGroup: /system.slice/duetcontrolserver.service
           └─1596 /opt/dsf/bin/DuetControlServer

Feb 11 20:28:52 duet3 systemd[1]: Starting Duet Control Server...
Feb 11 20:28:53 duet3 DuetControlServer[1596]: Duet Control Server v3.2.0
Feb 11 20:28:53 duet3 DuetControlServer[1596]: Written by Christian Hammacher for Duet3D
Feb 11 20:28:53 duet3 DuetControlServer[1596]: Licensed under the terms of the GNU Public License Version 3
~
~
~
~
~
~
~
~
~
~
~
~
~
~
~
~
~
~
~
~
~
~
~
~
~
~
~
~
~
~
~
~
~
~
~
~
lines 1-13/13 (END)

pi@duet3:~ $ journalctl -xeu duetcontrolserver.service -l debug
Failed to add match 'debug': Das Argument ist ungültig
pi@duet3:~ $ journalctl -xeu -l debug  duetcontrolserver.service
Failed to add match 'debug': Das Argument ist ungültig
pi@duet3:~ $ journalctl -l debug  duetcontrolserver.service
Failed to add match 'debug': Das Argument ist ungültig
pi@duet3:~ $ journalctl -l duetcontrolserver.service
Failed to add match 'duetcontrolserver.service': Das Argument ist ungültig
pi@duet3:~ $

wilriker

@taconite Sorry, I haven't been clear. So a more step-by-step instruction set.
First make sure to stop the duetcontrolserver service:

sudo systemctl stop duetcontrolserver

then run DuetControlServer manually with debug logging enabled:

sudo su - dsf -c "/opt/dsf/bin/DuetControlServer -l debug"

This will output everything directly to the console with logging in debug level.

taconite

This is the output

pi@duet3:~ $ sudo systemctl stop duetcontrolserver
pi@duet3:~ $ sudo su - dsf -c "/opt/dsf/bin/DuetControlServer -l debug"
This account is currently not available.
pi@duet3:~ $

wilriker

@taconite Interesting.

Please try

sudo /opt/dsf/bin/DuetControlServer -l debug

taconite

@wilriker said in Duet2 Wifi+Duex5 and SBC:

sudo /opt/dsf/bin/DuetControlServer -l debug

pi@duet3:~ $ sudo systemctl stop duetcontrolserver
pi@duet3:~ $ sudo /opt/dsf/bin/DuetControlServer -l debug
Duet Control Server v3.2.0
Written by Christian Hammacher for Duet3D
Licensed under the terms of the GNU Public License Version 3

[info] Settings loaded
[info] Environment initialized
[fatal] Could not connect to Duet (Board is not available (no header))
[debug] System.OperationCanceledException: Board is not available (no header)
   at DuetControlServer.SPI.DataTransfer.ExchangeHeader() in /home/christian/Due                                                                                                             t3D/DuetSoftwareFramework/src/DuetControlServer/SPI/DataTransfer.cs:line 1170
   at DuetControlServer.SPI.DataTransfer.PerformFullTransfer(Boolean connecting)                                                                                                              in /home/christian/Duet3D/DuetSoftwareFramework/src/DuetControlServer/SPI/DataT                                                                                                             ransfer.cs:line 162
   at DuetControlServer.SPI.DataTransfer.Init() in /home/christian/Duet3D/DuetSo                                                                                                             ftwareFramework/src/DuetControlServer/SPI/DataTransfer.cs:line 104
   at DuetControlServer.Program.Main(String[] args) in /home/christian/Duet3D/Du                                                                                                             etSoftwareFramework/src/DuetControlServer/Program.cs:line 113

again "no header" but I checked the wiring multiple times. Maybe there is something wrong with the small pcb (btw. I already checked for lifted pads and measured the connections)

wilriker

@taconite I never tested how good it works with the Wifi Module still present. In theory that should work but I only ever tested with a Duet 2 Ethernet with Ethernet module removed. So it's possible that there is an issue with that.

taconite

@taconite said in Duet2 Wifi+Duex5 and SBC:

(https://forum.duet3d.com/topic/17203/duet-2-ethernet-and-sbc)

mh okay I thought I read in this topic ((https://forum.duet3d.com/topic/17203/duet-2-ethernet-and-sbc)) that s.o. made it work with the wifi module still present. BTW the LED of the wifi module is not flashing and I installed the duet wifi server fw that came with 3.2 before I moved to the SBC duet FW

wilriker

@taconite Yes, I think another user got it working with Wifi module still installed. And also for D2SBC build the pin that enables Wifi module is explicitly held low on start-up such that the module does not start but still it being present could lead to signal detrimination.

Anyway, looking again at what you posted above, it says that there is no header. This at least means that the transfer ready pin is working correctly now. You said you have checked the wiring multiple times but please check again that did not mix-up MISO and MOSI - and in doubt try switching them on the RPi GPIO header.

taconite

@wilriker

okay seems like a little progress! Wiring was fine but I decided to replace the "little" board.
I get a connection but it seems unstable
and now I get something:

[debug] Connection to Linux established!

[warn] Restarting transfer because a bad header response was received (0xffffff01)
[debug] Updated key limits
[warn] Restarting transfer because a bad data response was received (0x0000006c)
[warn] Restarting transfer because a bad data response was received (0x00000022)
[warn] Bad data CRC32 (expected 0xd6ae08fc, got 0x0400dced)
[debug] Requesting update of key boards, seq 0 -> 0
[warn] Restarting transfer because a bad data response was received (0x0000006c)
[debug] Updated key boards
[debug] Requesting update of key fans, seq 0 -> 0
[debug] Updated key fans
[warn] Restarting transfer because a bad header response was received (0xffffff01)
[debug] Requesting update of key heat, seq 0 -> 0
[warn] Restarting transfer because a bad header response was received (0xffffff01)
[warn] Restarting transfer because a bad header response was received (0x000000fe)
[debug] Updated key heat
[debug] Requesting update of key inputs, seq 0 -> 0
[warn] Restarting transfer because a bad data response was received (0x00000200)
[debug] Connection to Linux established!

[warn] Restarting transfer because a bad data response was received (0xffffff6f)
[debug] Updated key limits
[warn] Restarting transfer because a bad data response was received (0x00000400)
[warn] Restarting transfer because a bad header response was received (0xffffff01)
[debug] Connection to Linux established!

[warn] Restarting transfer because a bad header response was received (0xffffff01)
[warn] Restarting transfer because a bad header response was received (0x000000fe)
[warn] Restarting transfer because a bad header response was received (0x000000fe)
[warn] Restarting transfer because a bad header response was received (0x000000fe)
[warn] Restarting transfer because a bad header response was received (0x000000fe)
[warn] Restarting transfer because a bad header response was received (0x10f001fe)
[warn] Restarting transfer because a bad header response was received (0xffffff01)
[warn] Restarting transfer because a bad header response was received (0x000000fe)
[warn] Restarting transfer because a bad header response was received (0x000000fe)
[warn] Restarting transfer because a bad header response was received (0x000000fe)
[warn] Restarting transfer because a bad header response was received (0x000000fe)
[warn] Restarting transfer because a bad header response was received (0x000000fe)
[warn] Restarting transfer because a bad header response was received (0x00000140)
[warn] Bad data CRC32 (expected 0x6b118bf2, got 0x9427c9a1)
[warn] Restarting transfer because a bad data response was received (0xffffff65)
[debug] Connection to Linux established!

[warn] Restarting transfer because a bad data response was received (0x00001000)
[warn] Bad header CRC32 (expected 0xffffffff, got 0x1523f9ff)
[warn] Note: RepRapFirmware didn't receive valid data either (code 0x0004005f)
[warn] Restarting transfer because the number of maximum retries has been exceeded
[debug] Connection to Linux established!

[debug] Updated key limits
[warn] Bad data CRC32 (expected 0x59fc6687, got 0x0af43244)
[warn] Bad data CRC32 (expected 0x59fc6687, got 0xa820c52f)
[warn] Bad data CRC32 (expected 0x59fc6687, got 0xfad539a4)
[warn] Restarting transfer because the number of maximum retries has been exceeded
[warn] Restarting transfer because a bad data response was received (0x00200020)
[debug] Connection to Linux established!

taconite

@wilriker Is there something I can do to stabilize the connection?

wilriker

@taconite You could try reducing SPI frequency in /opt/dsf/conf/config.json

taconite

@wilriker it is already lowered to 2MHz