[3.4b3] Homing coupler does not work correctly

Diamondback

@t3p3tony Hm, no luck Still the same broken behavior with no M574 command for C. Maybe there is some default that applies instead?

T3P3Tony

@diamondback what does sending:

M574

With no arguments report?

Also try sending
G1 H2 C-300 F10000
And
G1 H2 C300 F10000

From the console

Diamondback

@t3p3tony said in [3.4b3+] Homing coupler does not work correctly:

M574

It reports nothing for C when nothing is set, so that's good.

	M574
Endstop configuration:
X: high end switch connected to pin xstop
Y: low end switch connected to pin ystop
Z: none
C: none

The G1 commands also work on their own, it's literally just within the homeall file where things go south...

dc42

@diamondback I am having problems homing the coupler on my TC right now, which I am investigating.

dc42

@diamondback please can you try adding parameter R0 to the M584 command that creates the C axis, and reboot.

I am still investigating, however last time I had this problem, it disappeared when I rebooted the controller. So I don't want to do anything that needs a reset, such as trying this M584 change myself.

Diamondback

@dc42 Didn't change anything, with R0 it still homes incorrectly.

dc42

@diamondback please try the binary at https://www.dropbox.com/sh/xeu30go894p9xbs/AAB5U66SH4wbjKIcX0snUbKxa?dl=0. Caution if you are using stall homing: the stall sensitivity on Duet 3 MB6HC changed in release 3.4.0beta5 but has been reverted in this release.

Diamondback

@dc42 I'm afraid I'll need a Duet 2 (Wifi) binary.

dc42

@diamondback I've put one at https://www.dropbox.com/sh/xeu30go894p9xbs/AAB5U66SH4wbjKIcX0snUbKxa?dl=0.

Diamondback

@dc42 I'm carefully optimistic, I couldn't get it to do weird things yet, before it basically failed every single time. Will continue testing...

Diamondback

@dc42 Still no issues with the new binary, seems like you might have found the culprit

dc42

@diamondback thanks, I'll mark this thread as solved.

Diamondback

@dc42 I've been running the firmware build you posted here ever since you did, and no more issues with the coupler! Definitely seems solved!

Diamondback

@dc42 I have good news and bad news I just upgraded my entire TC from Duet 2 to Duet 3 hardware and also to 3.5b1.
The good news is that this issue did not happen anymore at all with the duet 2 hardware (on 3.4.5), however, now with Duet 3 on 3.5b1, it happens again I haven't tested Duet 3 on 3.4.5, so not sure if it's Duet 3 specific or 3.5b1...

Any ideas what could have caused this regression?

dc42

@Diamondback please check whether the symptom is exactly the same as you describes at the start of this thread, and start a new thread with any appropriate amendments. Post the config.g and relevant homing files for your new configuration.

Diamondback

@dc42 Hm, this is very odd timing.... It turns out, the grub screw of my coupler was loose and that's what made it look like it wasn't homing correctly... Weird that this happens right after a firmware upgrade

Not 100% sure this was the only issue, but I will assume so for now, sorry for the confusion.

dc42

@Diamondback thanks for the update! Over the years I have seen many instances of faults being blamed on a software update, when in reality something else just happened to go wrong at the same time.

Diamondback

@dc42 Yea same... I hate bad timing...

Diamondback

@dc42 As it turns out, this was not "just" a loose coupler and bad timing. I kept getting this issue even after fixing the coupler shaft grub screw.

It appears that the switch from Duet 2 Wifi to Duet 3 6HC for the coupler stepper had an impact on its torque. With the same settings in config.g it now stalled from time to time during homing before it reached its hard endstop (where it's supposed to stall).

I think I fixed it by increasing the current a little, at least I haven't seen the issue again afterwards.

Is a change of output torque expected when changing the stepper driver even if it's using the same settings?

dc42

@Diamondback said in [3.4b3] Homing coupler does not work correctly:

Is a change of output torque expected when changing the stepper driver even if it's using the same settings?

That's entirely possible, because the current can't be set to exactly the value you requested, as only certain values are available. RRF gives you the closest available lower current unless the next higher available current is only slightly higher. The TMC2160 or 5160 drivers provide finer current control than the TMC2209 at low currents.