T0 Trigger

that guy E

@that-guy-e
Testing this further with I am not seeing the board reset after the T0. The system is stuck in at halted.

I don't recall seeing this scenario with a duet2 running RRF 2.X

that guy E

Bump for visibility.

that guy E

Thought I had a work around setting up a customer trigger, but this will not work as it needs the current move to be completed. This is problematic if an estop was to occur and reset during a long move such as a home.

Is there any way to have a trigger or command be executed prior to the current moving being completed?

Phaedrux

What exactly do you mean by trigger a T0?

Can you share the gcode commands you're using?

It's not possible to interupt a move currently in progress by default. There is a work around for this where you configure it to break long moves into smaller sections allowing the interupt to happen sooner.

See M669 S and T parameters https://duet3d.dozuki.com/Wiki/M669

that guy E

@phaedrux

thanks for the suggestion, I'll try the additional M669 parameters. I had not seen them.

By T0 I was referring to the external trigger for an estop.

M581 T0 P1 S1 R0

This will register and halt move immediately but i get ground and over temp errors for all drives. the board then waits for a reset. With the Duet2 it would reset itself.

Phaedrux

@that-guy-e said in T0 Trigger:

i get ground and over temp errors for all drives.

That can be expected during an estop and isn't a sign of actual problems. It could be that they were suppressed before or didn't get displayed in time.

that guy E

@phaedrux

Is there a way to suppress messages?

What about restarting the board. With my duet2 running RRF 2.X it will restart after the estop condition automatically. With the duet3 running 3.3 it will stay halted till I power cycle or hit the onscreen stop.

that guy E

@phaedrux

Unfortunately it does not look like the segmentation is applied to a G1 move with an H1 parameter included. If I attempt a trigger during this move it will not register till the move is complete.

It looks like M581 T0 is the only way to interrupt this style move, which brings me back to my original issue of being in an unrecoverable state.

Maybe I can use the manual reset pins.

Phaedrux

Checking with @dc42

that guy E

@phaedrux

I switched up my wiring and think I found a work around. The estop button now kills 24V and a second contact on the estop makes the external board reset pins.

Preliminary test looks like that should work. Only concern long term would be corrupting an sd card on a read/write operation.

dc42

@that-guy-e currently a T0 trigger will halt the machine, not reset it. This guards against commands received via USB or elsewhere causing further motion. When the board is halted, it will still respond to M999 reset requests from DWC.

that guy E

@dc42

On a duet2 running RRF 2.x the board would reboot after an estop trigger without issuing a M999.

I think my current approach using the reset input on the duet3 should work. And I also don't see all of the driver errors like I do with m112.

If I was to upgrade an duet2 to RRF3.3 with a duet expansion how would I reset the board via an external input? If memory serves the reset pins on the duet2 are in the 50 pin.