New firmware 1.21RC3 available

Mike

I'm getting "Error: Attempt to seek on a non-open file." upon trying to print files from slic3r PE on 1.21RC3. The same files work OK on 1.20.

DjDemonD

I'm getting the same error on my corexy.

M122 - https://1drv.ms/t/s!Apv79JfGbPIwg5QsmYhl9eYy6yVTZg

when I try to downgrade to 1.21 rc2 I get

M997 S0
Error: M997: Firmware binary "Duet2CombinedFirmware.bin" not found

I managed to restore RC2 after hitting erase and flashing it using SAM-BA.
Working normally now, but still using wifiserver rc3, DWC rc3

Mike

Same, the error pops on a CoreXY machine. As for reverting to an older version… Just rename the old firmware to Duet2CombinedFirmware.bin and proceed as usual. Worked for me.

DjDemonD

I tried that, it didn't work, perhaps I didn't get the filename exactly right but I copy and paste it knowing it would need to be.

tjb1

@kraegar:

Kind of odd. If I power on, go to the gcode console, and send G1 Z-10 S2 it'll move as expected. If I send the same thing again, though, no movement. But if I change the Z value to something else, like G1 Z-20 S2, then it'll move as expected.

To further confuse me, now it's moving the opposite direction I should after a few more Z moves before homing.

9:44:10 AMG1 Z-20 S2
9:44:02 AMG1 Z-30 S2
9:43:53 AMG1 Z-40 S2
9:43:44 AMG1 Z-20 S2
9:43:39 AMG1 Z-20 S2

The Z-30 and Z-20 both moved the bed in positive Z instead of negative.

Those commands are being sent as absolute, once you move to Z-10, it can't go anywhere on the second Z-10 because it's already there. Same thing with the second issue, you are already at Z-40 and going back in the positive direction to Z-30 and Z-20, both are 10+ moves.

garyd9

I don't know if this was fixed in RC3 or not, but I just noticed the following issue in RC2:

When setting up a z-probe with P5 (It's a BLtouch) A3 (probe each location up to 3 times) and B1 (turn off bed heater while probing), the heater gets turned off for the FIRST of multi-tap probes, but not subsequent multi-tap probes.

So, assuming a M558 of:

M558 P5 X0 Y0 Z1 H5 F100 T4000 A3 B1

and a bed.g containing the following:

G30 P0 X1.1 Y-69.2 Z-99999
G30 P1 X-50 Y65.5 Z-99999
G30 P2 X49 Y65.5 Z-99999 S3

(and the manual levelling assistant configured.)

In each of the above probes, the heater is turned off for the FIRST test in each location, but the heater is turned back on (and stays on) for the SECOND probe. (My probes are apparently within 0.03 of each other, as I don't see a third probe in any location.)

kraegar

@tjb1:

@kraegar:

Kind of odd. If I power on, go to the gcode console, and send G1 Z-10 S2 it'll move as expected. If I send the same thing again, though, no movement. But if I change the Z value to something else, like G1 Z-20 S2, then it'll move as expected.

To further confuse me, now it's moving the opposite direction I should after a few more Z moves before homing.

9:44:10 AMG1 Z-20 S2
9:44:02 AMG1 Z-30 S2
9:43:53 AMG1 Z-40 S2
9:43:44 AMG1 Z-20 S2
9:43:39 AMG1 Z-20 S2

The Z-30 and Z-20 both moved the bed in positive Z instead of negative.

Those commands are being sent as absolute, once you move to Z-10, it can't go anywhere on the second Z-10 because it's already there. Same thing with the second issue, you are already at Z-40 and going back in the positive direction to Z-30 and Z-20, both are 10+ moves.

Yep, I realized that once I had some more coffee.

Amilious

I think that I might have found a bug.
When doing the command
G1 X4 Y4 F200 S2;
only the X-axis moves

I am using this in my homeall.g

Here is my old homeall.g with added S2 does not work

[[language]]
G91				               ; Relative Positioning
G1 Z5 F6000 S2			; Lower Z
G1 X4 Y4 F600 S2             	; move away from the endstops
G1 S1 X-305 Y-305 F3000   	; course home X or Y
G1 S1 X-305               	        ; course home X
G1 S1 Y-305               	        ; course home Y
G1 X4 Y4 F600 S2             	; move away from the endstops
G1 S1 X-10                	        ; fine home X
G1 S1 Y-10                	        ; fine home Y
G90				                ; Absolute positioning
G1 X8 Y30 F5000 S2		; Move to first probe point
T0 				                ; select tool	
M280 P3 S160 I1			; release Z probe alarm
G1 Z5 F200 S2 			; lower bed
G4 P500 			                ; wait for the bed to lower
M401				        ; deploy the probe
G30 				                ; calibrate Z-axis
M402				        ; Retract Probe
G1 x0 y0 F5000			; Move to 0,0
G1 Z0 F200			        ; Move to Z 0

Amilious

@DjDemonD:

I'm getting the same error on my corexy.

M122 - https://1drv.ms/t/s!Apv79JfGbPIwg5QsmYhl9eYy6yVTZg

when I try to downgrade to 1.21 rc2 I get
M997 S0
Error: M997: Firmware binary "Duet2CombinedFirmware.bin" not found
I managed to restore RC2 after hitting erase and flashing it using SAM-BA.
Working normally now, but still using wifiserver rc3, DWC rc3

Maybe try renaming the old firmware that you are trying to downgrade to "Duet2CombinedFirmware.bin". The newer versions look for that file name, the same way you had to rename the Duet2CombinedFirmware.bin to DuetWiFiFirmware.bin when upgrading the first time.

dc42

When upgrading and downgrading between firmware versions in which the expected name of the firmware binary has changed, you need to take account of:

1. The binary name that the current installed firmware expects. Firmware 1.21RC3 expects Duet2CombinedFirmware.bin. Firmware 1.21RC2 on the Duet Ethernet expects DuetWiFiFirmware.bin. Firmware 1.21RC1 and earlier expect DuetWiFiFirmware.bin or DuetEthernetFirmware.bin depending on which board you have.

2. The binary name that the currently-installed version of DuetWebControl expects. DWC 1.21RC4 expects Duet2CombinedFirmware.bin. Older versions of DWC expect DuetWiFiFirmware.bin or DuetEthernetFirmware.bin depending on which board you have.

What always works is:

upload the firmware binary you want to install;
use the System Editor page of DWC to rename it to the name expected by the currently-installed firmware;
send M997 to install it.

DjDemonD

Okay, well that's simple enough. That was though just a minor issue, its the fact that RC3 doesn't seem to work on corexy that was the main event for me.

dc42

@DjDemonD:

Okay, well that's simple enough. That was though just a minor issue, its the fact that RC3 doesn't seem to work on corexy that was the main event for me.

Please provide more details of why RC3 didn't work for you, along with any relevant GCode files.

kraegar

It's working with my CoreXY OK.

(I'm chasing some print artifacts, but they exist on 1.20 as well, so I don't think it's FW related)

Mike

In my case it was "Error: Attempt to seek on a non-open file." when trying to print anything sliced on Slic3r PE on my CoreXY. Reverted to the old firmware and finished the print without issues. I'll run a couple more tests on both my delta and the CoreXY in the evening.

kraegar

I get that right at the end of files from Cura. Not at all from S3D. But it prints the full print for Cura.

DjDemonD

@dc42:

@DjDemonD:

Okay, well that's simple enough. That was though just a minor issue, its the fact that RC3 doesn't seem to work on corexy that was the main event for me.

Please provide more details of why RC3 didn't work for you, along with any relevant GCode files.

Hi David sorry scroll up a bit I provided a report and an M122. Corexy and attempting to print a SlicerPE file, so same issue others have had, downgrading just FW fixed it. Deltas working perfectly on rc3.

resam

I'm also running a CoreXY - didn't experience any issues with RC3 so far, about 5 prints or so successfully finished.

DC42: I think it would make sense to add the new debug information about SD cluster size to M122 as well.

dc42

@Mike:

In my case it was "Error: Attempt to seek on a non-open file." when trying to print anything sliced on Slic3r PE on my CoreXY. Reverted to the old firmware and finished the print without issues. I'll run a couple more tests on both my delta and the CoreXY in the evening.

Thanks. Please can you confirm or otherwise:

1. The fault only occurs when you try to print a file from SD card. In particular, just listing the files on the DWC GCode Files page doesn't provoke the problem;
2. The fault only occurs if the file was sliced by slic3r PE;
3. All files sliced with this version of slicer PE exhibit this issue when you try to print them.

Please supply a small test file that demonstrates the problem.

appjaws

I have had to revert to rc2
I need to move the hotend to the center of the bed prior to Z0.
I also have a startup procedure in my start gcode on S3D. Even though the machine had all axis zero'd the print would not start. So I disabled the homeall part of the startup but because I use M32 to level the bed using 3 motors and that includes homeall nothing happened. I then noticed that the heaters had been turned off due to a power reset!!!!

Why are we introducing a procedure that Z has to be zero'd before and other axis can move? This may well be good for Delta printers but is no good for my coreXYUV machine. This seems a backward step.
Would it be possible for a user option to switch this on or off?
Sorry for the rant, feeling a bit dissapointed

dc42

Z does not have to be homed or zeroed before the other axes can move. Did you read the upgrade notes?