Attempting to extrude with no tool selected - During Print

dc42

@jaurand, does your system include a PanelDue, and if so, how long is the cable and what baud rate are you using for its connection?

jaurand

@dc42 It does include a Paneldue, baudrate is set to 57600 on both the panel and in the config file using M575.

At work at the moment so not able to be exact on cable length but its not super long, maybe 12 to 16 inches.

dc42

Please send M575 P1 to your Duet from the console and check that the response includes "requires checksum".

jaurand

@dc42 said in Attempting to extrude with no tool selected - During Print:

M575 P1

Yep requires check sum

M575 P1
Channel 1: baud rate 57600, requires checksum

dc42

We had situations a few years ago in which commands from PanelDue were getting corrupted and causing printing to stop. That's why we set the port to require a valid checksum. Unfortunately it's only and 8-bit checksum, so errors can still get through. In case this is causing the problem, I suggest you test whether the problem goes away if you disconnect the PanelDue. If it does, then check the PanelDue cable for a bad connection. The resistance of the cable must be not more than 0.1 ohm per conductor.

jaurand

@dc42 i was wondering to myself if it could be the paneldue. I do get some checksum errors still passed on. I'll try out next week a few more prints to see if that it and test the resistance.

dc42

@jaurand said in Attempting to extrude with no tool selected - During Print:

@dc42 i was wondering to myself if it could be the paneldue. I do get some checksum errors still passed on. I'll try out next week a few more prints to see if that it and test the resistance.

Checksum errors suggest that there is a problem with the PanelDue to Duet link. For every 255 checksum errors reported, there is probably 1 case of data corruption that isn't detected and may cause problems.

What microstepping settings are you using? It's possible the use of high microstepping could increase interrupt latency and give rise to data reception errors. the 2.03 firmware series has some changes to help avoid this happening. So you could upgrade your firmware to 2.03RC4 and see whether the checksum errors reduce or go away.

jaurand

@dc42 Tested connectors on paneldue and all are below 0.1 ohms. Currently unplugged paneldue and printing will see if any issues arise.

I am only running X16 microstepping with interpolation at the moment as my motors are stock with the printer and I did not have very detailed information regarding the specs.

jaurand

@dc42 No issues with the print from tool selection with paneldue unplugged. I'll look into updating the firmware next and determine if this reduces the checksum errors

jaurand

@dc42 said in Attempting to extrude with no tool selected - During Print:

So you could upgrade your firmware to 2.03RC4 and see whether the checksum errors reduce or go away.

upgraded to 2.03RC5 and still getting checksum errors

dc42

What sort of cable do you have between the PanelDue and the Duet? Is it running close to stepper motor wires for any appreciable distance?

jaurand

@dc42 said in Attempting to extrude with no tool selected - During Print:

What sort of cable do you have between the PanelDue and the Duet? Is it running close to stepper motor wires for any appreciable distance?

I'm running the 4 pin wire. The cable is not close to any stepper motor wires

dc42

It sounds to me like a signal quality issue, but it could possibly be caused by an overloaded MCU. Are you using unusually high microstepping?

jaurand

@dc42 just x16 with interpolation

dc42

I am surprised you are getting checksum errors. I don't recall seeing see any reported on my delta for over a year, and the PanelDue cable is about 1.3m long.

You could try reducing the baud rate to 38400 or 19200 at both ends, to see if the checksum errors go away.

jaurand

@dc42 dropped the baud rate and still had check sum errors. also to note I had lost the ability to baby step with upgrading firmware to 2.03RC5 so i have gone back to 2.02 and able to baby step there

Scachi

@jaurand I had a lot of checksum errors too. I tried to get rid of this by mounting the duet wifi board to a different location on my printer, changing the baud rate, using different cables for the paneldue, mounting the paneldue to a different location.

Searching the forum/wiki I found that twisting the wires of the endstop cables can fix it and It looks like this did the trick. No more checksum errors here.
My x&y endstop cables are very close to (directly above of) one of my stepper motors: https://www.thingiverse.com/thing:3667947

THaala

Hello,

same problems on my printer.
After Update to 2.03 and Panelue 1.23 i got a lot of checksum errors. I didnt have seen this before (2.0 beta) with 115200 Baud.

Reducing raudrate to 9600 reduces the amout of errors a bit but not much.
Software <= 2.0 beta have never shown such checksum errors.

The problem with "attemepting to extrude with no tool selected" during print i have seen on time destroying a 13 hours print.

BR,
Thilo

dc42

You should not have such a high error rate. One cause could be that the PanelDue cables are running along the printer frame, and you haven't grounded both the printer frame and the negative output form the power supply. That causes the printer frame to carry noise from stepper motors and heater switching, which is coupled into the cable.

I have a 1.5m cable between the Duet and PanelDue on my delta, and I get zero error reports.

Phaedrux

With PanelDue firmware 1.23 I started to get some checksum errors as well. I had previously noticed some screen flicker. So I decided to replace the cable with a new recrimped one and now everything is fine again. No screen flicker, no checksum errors. My frame is grounded as well as the DC- of the PSU.

Perhaps some of the newer features in 1.23 are exacerbating existing poor cabling.