By the way, I am a huge fan of the smart effecter as well and would recommend the 288mm mag rails.

I didn't cut the wire to put a resistor. I am going to try your settings right now.
This board is a pain but I'm sure I will love it when I get it right. I have been trying so hard
for 3 weeks now. I get so mad I want to throw it in the lake. Just kidding. I love it and wish I could get it going.

Thank you

@dc42 Now I know what happened when M905 gave me "Clock has not been set":

I updated config.g through the web interface. The UI asked me whether I want to restart to make the new settings effective. I chose to do so. Then DWC reconnected somehow in the background automatically (or never got disconnected in the first place because restart is so quck?). Now as long as I do not disconnect/reconnect the time will not be set anew.

@dc42
I cant see any conflicts. one thing I did notice was if I changed the speed in the first X move in homex.g to 180 from 1800 mm/m to give me time to press the switch by hand while the axes was moving towards home the motor sped up after pressing it so its seeing the switch
I was able to determine that the speed at which the motor sped up to is the programmed speed in the next G1, X move which is "; go back a few mm"

next I copied and pasted the homex example from dozuki doc. "Configuring RepRapFirmware for a Cartesian printer" and it slowed down after hitting the switch manually on its way to home which is how it is written.
G91 ; relative mode
G1 Z4 S2 F200 ; raise head to avoid dragging nozzle over the bed
G1 X-240 F3000 S1 ; move up to 240mm in the -X direction, stopping if the homing switch is triggered
G1 X4 F600 ; move slowly 4mm in the +X direction
G1 X-10 S1 ; move slowly 10mm in the -X direction, stopping at the homing switch
G1 Z-4 S2 F200 ; lower the head again
G90 ; back to absolute mode

seems the S1 is not producing the desired effect. Before this test with homex.g I have not changed any other FW files except for config.g. Sorry to be a PITA

@whosrdaddy said in Extruder sometimes cuts out in middle of long prints:

https://forum.duet3d.com/topic/3483/extruder-motor-stops-during-print

In 10 hrs of printing since adding grounding the extruder motor body (on a PG35L), with two prints, I haven't seen this issue recur yet. This isn't proof of a fix, but it's highly encouraging, since the MTBF seemed on the order of a few hours previously.

I used a wire that is tucked between the outer shell and the heatsink, and connected to the power supply ground.

Excited to be able to try other stuff now - updating to Duet firmware 2.0, trying out other slicers, etc!

By default, firmware 1.21 does not allow normal axis movements (including jogging) before that axis is homed. See the upgrade notes for 1.21 for details and how to revert to the old behaviour if you need to.

Stack overflow messages generally mean that you have created a recursive macro. For example, you may have a G31 command in homeall.g and a G28 command in bed.g.

See section 5 in the TMC2660 datasheet at https://www.trinamic.com/fileadmin/assets/Products/ICs_Documents/TMC2660_datasheet.pdf. The M122 report includes the highest and lowest SG register values since M122 was last run, to help you find the correct coolstep setting.

I have looked again at the implementation of M502. Currently it resets only those parameters that are normally overridden in config-override.g. However, I think it would be more compatible with how other firmwares behave if it ran almost all of config.g. So in the next release candidate it will run all commands in config.g except for network-related commands, T commands, and M501 commands.

I found the problem, I had G32 in the bed.g instead of G29. Thank you for the help

@deckingman said in Bondtech (or similar) extruders:

Update on this.

Thanks for the update and the detailed blog post and video. I've had a few retraction related issues on my Kossel Mini with Titan extruder, and I've also had to disassemble it twice to get things to line up properly, and to deal with the bad clicking issues. This in under 200 hours of use. I assumed 90% of those issues were my fault, but after reading this thread, now I'm not so sure. I might try the Bondtech BMG on my next build -- although replacing the Titan on this Kossel Mini might also make sense.

John

nevermind 🙂 By setting the max temp using M143 is the solution

Try adding more bed size in the Homeall.g for example here is a modified section of yours.

G1 Y-220 F1800 S1
G1 X-220 F1800 S1 ; Move towards X and Y axis endstops (first pass)
G1 X5 Y5 F6000 ; Go back a few mm
G1 X-220 Y-220 F360 S1 ; Move slowly to axis endstops once more (second pass)

Even though your bed size is 195mm in some conditions it may take more than 195mm of travel to make it to the endstop while homing.

I noticed that in your homex.g you gave the X position 250mm of travel during homing of just X, you can use that number in homeall also if you feel like it.

I can't think of any changes that would account for that behaviour. My guess is that you have the Z acceleration configured too high in M203, or the Z speed too high in M201, or the Z motor current set too low.

Looks promising. The file name one th git hub doesn't have the revision number that the dev releases from the website have: https://dl.slic3r.org/dev/win/

That said the last dev release and the 1.3.0 file on github seem to have been posted on the same day. That said a final revision would probably be made to go through some internal test sheets before being released as a final version.

Tested, it works. Thanks DC!

For missed step detection to work, you need to keep the speed below the "low slip angle" maximum speed.

The M203 setting is the never-exceed-speed. So it should be higher than your printing speed and travel speed settings in slic3r. A typical value for a delta would be 12000mm/min i.e. 200mm/sec. Don't set it so high that your printer starts to skip steps when you command movement at maximum speed.

Thanks for the input. I'll leave the G32 code in my slicer.

I've made a little plug with a 100k resistor that I use when only using 1 hotend and it works perfectly well (temp is reported at 24 .5 degrees).
Thanks for the adivce 🙂

Thanks. I will have to check with @W3DRK.

Ben