Thank you so much to everyone who helped me through this upgrade; especially @Phaedrux who spent a disproportionate amount of time on my problems. You guys are awesome!

I've learned a lot going through this process and have lost a lot of the trepidation I felt when I used to think of making changes to the firmware. Now I have an up-to-date firmware and future upgrades should be easy.

I PID tuned the bed and hotend heaters (that was so easy and the Duet calibration guide is clear and very well done), adjusted the homez.g and homeall.g files a bit to make Z drop a bit after hitting the endstop (I home to the top), and I am making a few tweaks to the fans. There might be more but I did make a nice short print already.

Tomorrow I'll try adjusting pressure advance again and start tweaking other settings. And I'll make backups.

Thank you all again! I'm so happy to be back printing.

Just a follow up; since the upgrade I've printed steadily and have yet to experience a disconnect from DWC. That appears to be fixed (crossing my fingers).

Well I give up trying to run the code in my slicer; I can't get it to work in either PrusaSlicer or SuperSlicer.

But I could send commands from the console and pressure advance does work when added to my filament g-code so I'm happy.

I can say it wasn't working for a long time because the extruder stepper shaft moves differently now than before; I can see the stepper rotation indicator move with slight twitches at the end of each path.

Thank you everyone for your suggestions!

@Norder Oh thank you! This is very good and even easier than picking a scrollbar (if one existed). Because normally I click a scrollbar before using arrow keys I didn't even think to try this because said scrollbar didn't exist. Your suggestion is quite perfect - thank you again.

@Phaedrux said in Idiot's Guide To Upgrading Duet Firmware:

Small world.

Wow, that's cool. It's surely is a small world.

@t3p3tony I'm getting one e-mail now. All is good. Best regards.

@droftarts said in Can't run pressure advance g-code:

Check you are using Gcode with D0 rather than D3.

Thanks. I have been using D0 after @tas clarified that.

At this point I've changed so many variables and settings, and used different slicers, etc., that I don't know what wasn't working, BUT now that code runs through the slicer and is working.

I ran a coarse test and then narrowed the range down for a second test with finer ranges and now have a pressure advance value that's working well with PET-G. I'll run the test again with PLA and I'm really eager to see how all my machine's new settings with work with Ninjaflex. I really can't believe how much pressure advance is helping already.

I have a direct drive extruder (homemade with 78mm from the center of the drive gear to the nozzle tip) but seem to require quite a high pressure advance of S0.650 (2.85mm PET-G). Here's a picture (sorry for the poor quality).

Even though I don't know why it wasn't working, it is now, so I can say the issue is solved. Thanks a lot to everyone who helped with this!

@Phaedrux said in Idiot's Guide To Upgrading Duet Firmware:

That sounds correct for your config.

Awesome! Thank you for checking that. Then this firmware upgrade is a success!

I'm printing now and can see some differences immediately. Infill is now connecting to perimeters; that was a thorny issue I could not solve for a long time but is much better now.

@tas It's a game of iteration and never being satisfied.

Here's another print that has improved markedly after this firmware upgrade. I'm not sure why but wonder if part of the reason is pressure advance is working and tuned, and my acceleration and speed change settings are higher than before? Regardless, I'm pleased.

Just a follow up; since the upgrade I've printed steadily and have yet to experience a disconnect from DWC. That appears to be fixed (crossing my fingers).

@Phaedrux said in Z Maximum and Z Position and Homing:

Nothing wrong with that.

@Phaedrux said in Z Maximum and Z Position and Homing:

That way you can position the Z axis however you want.

Thank you! It's so cool what the Duet can do. I'll be trying that for sure at some point.

Right now I have my machine set so I don't need any Z offset. Because of my disability I sometimes have to do things in non-conventional ways because I can't do things the normal way. To get the Z height I adjust the Z offset in the slicer and when the print shows a perfect first layer (or skirt thickness) I adjust the machine Z maximum by that amount and the result is 'close enough'.

The above assumes the machine Z maximum is already close.

I also tram the build plate, when needed, by printing dots in the center and corner of the bed and then adjust as needed. It's messy but at least I can do that without help.

@Phaedrux said in Z Maximum and Z Position and Homing:

With an accurate measurement of the Z0 position you shouldn't need any slicer offset.

The smallest moves the macro makes are 0.05mm (I'd like to know how to adjust that value) and I don't find 0.05mm fine enough to get a perfect first layer; particularly when printing tiny parts with very little contact area to the bed.

My Z leadscrews have an 8mm pitch but even with them 16x micro-stepping gives theoretical Z resolution of 0.02mm. I'm not sure what actual resolution the machine is capable of but I adjust my Z offset in 0.02mm amounts and it makes changes that give me very good first layer control.

Maybe I'm missing something with the macro? If I could jog the X axis down by 0.02mm it would be perfect. To avoid trying to see the nozzle touch the bed I'd put a 1mm thick steel ruler under the nozzle and jog down till I could feel the ruler just start to drag on the nozzle and then pull it out. Then I'd lower the nozzle 1mm and then move the axis to the top.

That is a really cool macro.

@dc42 said in Idiot's Guide To Upgrading Duet Firmware:

The maximum instantaneous speed settings are best left alone unless the machine stutters when printing circles.

Thank you.

Sorry, I mislead you. After upgrading I reduced some speeds and now want to try increasing them to as close to the defaults as possible.

My machine has heavy axes so bangs hard if the speeds are too high.

These are its current settings:

M566 X1000.00 Y1000.00 Z24.00 E3000.00             ; set maximum instantaneous speed changes (mm/min)
M203 X30000.00 Y30000.00 Z900.00 E2700.00          ; set maximum speeds (mm/min)
M201 X1000.00 Y1000.00 Z200.00 E3000.00            ; set accelerations (mm/s^2)

@droftarts said in Can't run pressure advance g-code:

Check you are using Gcode with D0 rather than D3.

Thanks. I have been using D0 after @tas clarified that.

At this point I've changed so many variables and settings, and used different slicers, etc., that I don't know what wasn't working, BUT now that code runs through the slicer and is working.

I ran a coarse test and then narrowed the range down for a second test with finer ranges and now have a pressure advance value that's working well with PET-G. I'll run the test again with PLA and I'm really eager to see how all my machine's new settings with work with Ninjaflex. I really can't believe how much pressure advance is helping already.

I have a direct drive extruder (homemade with 78mm from the center of the drive gear to the nozzle tip) but seem to require quite a high pressure advance of S0.650 (2.85mm PET-G). Here's a picture (sorry for the poor quality).

Even though I don't know why it wasn't working, it is now, so I can say the issue is solved. Thanks a lot to everyone who helped with this!

@Phaedrux No worries! Thank you.

Actually the value was written to config-override.g and is, I believe, working.

Interestingly the macro says my Z maximum is ~0.25mm more than the Z maximum I've been using so without adjusting my slicer Z Offset the first layer was too squished. This could be due to how I ran the test.

I measured a piece of thick-ish paper that was on hand and placed it under the nozzle and dropped the nozzle till I could slide the paper out without too much trouble. Then I dropped the nozzle the thickness of the paper before putting the axis back to the top.

I'll run the macro again.

Everything seemed to run perfectly till after the X axis moved to the top and touched the limit switch and an error message popped up saying the process took too long and to see the message in the console:

@Phaedrux said in Idiot's Guide To Upgrading Duet Firmware:

... some things are probably a bit faster which can help quality as you don't want the print head hanging around oozing out plastic when it shouldn't be

The settings are definitely faster. I'll try increasing acceleration and maximum instantaneous speed changes as much as possible before the machine complains.

@tas said in Idiot's Guide To Upgrading Duet Firmware:

I have found that every upgrade brings some improvement to print quality.
Of course the slicers have improved too.

For sure; it's exciting to be along for the ride!