Firmware 2.02RC2 released

[genghisnico13]

I got the error "Error: motor phase A disconnected on drivers 1
Error: motor phase B disconnected on drivers 1" when pushing my motors. It doesn't seem to affect anything.

[Phaedrux]

I've done a few prints and things seem ok on my CoreXY.

[dc42]

@genghisnico13 said in Firmware 2.02RC2 released:

I got the error "Error: motor phase A disconnected on drivers 1
Error: motor phase B disconnected on drivers 1" when pushing my motors. It doesn't seem to affect anything.

Do you get that message frequently? Please run M115 to check that you really are running 2.02RC2, not 2.02RC1.

[genghisnico13]

@dc42 I have the RC2:
FIRMWARE_NAME: RepRapFirmware for Duet 2 Maestro FIRMWARE_VERSION: 2.02RC2(RTOS) ELECTRONICS: Duet Maestro 1.0 FIRMWARE_DATE: 2018-09-07b2
I only saw it once (since the update to RC2), but I haven't been printing these past few days, I still have a few tests and calibrations pending before I start printing.
I'll try to see if I can find a way to trigger the error, if I can I will post here.

[genghisnico13]

Well, that didn't take long.
Ive been writing a script to test the limits of my printer travel acceleration, with this Gcode I can trigger the error, it isn't 100% effective, but after running it a few times the error appears.
0_1536974933035_YMax7000mms2 steps200.g
This Gcode was designed for Cartesian printers and needs to have 200mm of movement in the Y direction with a MIN limit switch.

It basically moves the bed to position 200 at a known safe speed(50mm/s I believe), tweaks the acceleration(to the value to test) and moves to position 20 at 300mm/s, then sets the acceleration to a known safe value and goes to position 0, then moves to -0.1mm(at really low speed which could be the trigger based on what I read) to make sure the endstop is triggered and sends an M119 and loops again with a different acceleration.
Since we don't have conditional Gcode(I know that its going to take a lot of work to implement if ever), this is the best semi-automatic way to find the limits of my motors. In the console you can see the acceleration at which steps are lost:
10:13:03 PMY axis Acceleration 6600mm/s2: Endstops - X: at min stop, Y: not stopped, Z: not stopped, Z probe: not stopped
10:12:56 PMY axis Acceleration 6400mm/s2: Endstops - X: at min stop, Y: at min stop, Z: not stopped, Z probe: not stopped
10:12:49 PMY axis Acceleration 6200mm/s2: Endstops - X: at min stop, Y: at min stop, Z: not stopped, Z probe: not stopped
edit: the steps are set low(200mm/s2) specifically to make it long and trigger the error.

[dc42]

Thanks, this is now on my list to investigate. I probably won't have time to look at it until after TCT.

[genghisnico13]

@dc42 no problem, it's not a priority, as far as I can tell there are no adverse effects.

[genghisnico13]

@dc42 I discovered that the error happens more often in stealthchop, I've been testing a silent profile with M569 V0 and the disconnected phase error happens more often, all this tests are made in a CR-10 that has high inductance motors at 14V.

Just in case I reiterate: not a priority, it really doesn't bother me at all, I just wanted to make it easier for you to replicate the error.

[john]

Running dual motor Y-axis on Driver 1 & 2. Received "Error: motor phase B disconnected on drivers 1 2" today.

Duet 2 WiFi/Ethernet version 2.02RC2 (RTOS). It halted my operations because the parent thread received the unexpected message above from the GCode thread.

[dc42]

@sean said in Firmware 2.02RC2 released:

Running dual motor Y-axis on Driver 1 & 2. Received "Error: motor phase B disconnected on drivers 1 2" today.

Duet 2 WiFi/Ethernet version 2.02RC2 (RTOS). It halted my operations because the parent thread received the unexpected message above from the GCode thread.

What parent thread was that? RepRapFirmware takes no action when it detects a motor disconnect other than reporting it.

[kuhnikuehnast]

I get the same error of the "motor phase A disconnected on drivers 

It always happens if I start the auto bed compensation. (Prusa Style Printer, 2 Stepper Drivers for Z to adjust the axis).

Code of Bed.g:
M561 ; clear any bed transform

G30 P0 X20 Y20 H0 Z-99999
G30 P1 X200 Y20 H0 Z-99999 S2

If needed, I could also paste my config.g? (how do you get it to be shown as "source code"?

greetings kuhni

[john]

@dc42 We run things a little different than most users seem to. We have a parent python thread that sends GCommands to the Duet through pyserial(). When an messages is sent back from the Duet that is not "OK" or "Endstop Status: ..." the thread sending GCode stops (because presumably something unexpected happened).

[Catalin_RO]

@dc42 I had a quick look through the latest source files and found this rather strange looking piece of code in GCodes.cpp, in GCodes::DoArcMove

if (!seenXY && seenIJ)		// at least one of XY and IJ must be specified
{
	return  "G2/G3: missing parameter";
}

Some how the condition looks wrong, as G2/G3 with no parameter would pass through instead of being just a NoOperation.

[dc42]

Thanks, it should be "if (!(seenXY && seenIJ))" to match the comment. I will fix it in the next build.

[Russ Gries]

@dc42 said in Firmware 2.02RC2 released:

M115

yeah that's annoying

FIRMWARE_NAME: RepRapFirmware for Duet 2 WiFi/Ethernet FIRMWARE_VERSION: 2.02RC2(RTOS) ELECTRONICS: Duet WiFi 1.0 or 1.01 FIRMWARE_DATE: 2018-09-07b2

also, i have a stepper motor that only draws 40ma max. is this why i get this message or is this known error?

Just thought i would post it. thanks!
~Russ

[JoergS5]

@dc42 said in Firmware 2.02RC2 released:

Thanks, it should be "if (!(seenXY && seenIJ))" to match the comment. I will fix it in the next build.

I would take if(!(seenXY || seenIJ))
I am not a native english speaker, but at least one means one of the two or both?

[dc42]

@joergs5 said in Firmware 2.02RC2 released:

@dc42 said in Firmware 2.02RC2 released:

Thanks, it should be "if (!(seenXY && seenIJ))" to match the comment. I will fix it in the next build.

I would take if(!(seenXY || seenIJ))
I am not a native english speaker, but at least one means one of the two or both?

At least one of IJ is needed to define the arc centre, and at least one of XY is needed to define the end position. The comment could be clearer.

[JoergS5]

@dc42 Thank you for explaining.

[T3P3Tony]

@russ-gries try updating to the latest DWC

[kuhnikuehnast]

@T3P3Tony - same Issue here... But on the latest DWC:

Firmware Version: 2.02RC2(RTOS) (2018-09-07b2)
WiFi Server Version: 1.21
Web Interface Version: 1.22.3