Duet Maestro Freefall



  • I hate to come post here but I think you guys may have a better idea as to whether this is inherent to the Trinimac stepper chips or something that could be addressed in firmware or with a settings. current firmware 2.02beta1.

    The primary printer the Maestro is installed on is the new M3D Promega which has a... novel... belt design for the Z axis. When the motor over currents, rather than skipping a step the drivers seem to cut off and cause the motors to freewheel (not limited by being a generator somehow) and the bed crashes to the floor of the printer. This, at a minimum skips belts and requires a full bed re-calibration.

    Is it possible to just disable the steppers and go into a stop condition and let the natural force of the steppers let the bed drift to the bottom like it does when you power the machine off or is this a limitation of the stepper driver chips themselves?

    Thanks!


  • administrators

    Please explain what you mean by "When the motor over currents".



  • The current behavior is that when the bed is pushed up into the nozzle, I would expect this to overcome the motor torque and cause a skipped step. Instead the motor seems to reach current limit, the driver shuts off(?) and the the bed drops at a high rate of speed (gravity? Feels like faster than gravity) and crashes to the bottom. I could take the hit and provide a video if needed for diagnostics.



  • @leonmf said in Duet Maestro Freefall:

    Instead the motor seems to reach current limit, the driver shuts off(?) and the the bed drops at a high rate of speed (gravity? Feels like faster than gravity) and crashes to the bottom
    Hi,

    Is the movement in the axis so free the the bed will drop on it's own due to gravity?

    Could you post a picture of your machine?

    Thanks.

    Frederick



  • @fcwilt Yes, the bed is belt driven from a single Z axis motor. With the stepper disabled, the bed will slowly drop while acting as a generator. When the stall detection kicks in (I just learned this was a feature and I presume they may be using M915? I'm going to check when I get home) the motors are disconnected from the stepper and the bed falls like a rock. Given that it's a 15x15 glass bed with aluminum frame... this is undesirable behavior and will break things!

    https://store.printm3d.com/pages/promega

    I can post a picture later if you still want one.



  • @leonmf said in Duet Maestro Freefall:

    I can post a picture later if you still want one.

    Thanks much. That is a large and pricey printer. I guess for that kind of money that can provide a low friction setup.

    My FT5 moves smoothly but it slowly sinks down without crashing.

    Frederick



  • I hope this printer becomes a solid workhorse but, I'm afraid, it is still more of a beta solution and this particular problem is one of the more vexing ones!



  • Here are some videos highlighting the issue:

    Duet Power off free-fall:
    https://youtu.be/WCvHGXi6jxs

    Duet Power On Free-Fall
    https://youtu.be/7uG-6czPOVY

    Duet Bed Stall crash
    https://youtu.be/QH9c-w1Nbgg



  • Wow! I'd say something needs to be done about that.......



  • @leonmf OK. So what seems to be happening is that the bed is being driven in minus Z because those moves are very much faster than gravity with no power. So if it's being driven, then there must be something in your files telling it to move to Znnn at a feedrate (speed) of Fnnn. Might be an idea to post your home Z file, or at least take a look at it and see what is being commanded after the probe or stall detection or whatever method you use to home, has been triggered.

    Edit. And/or you have the steps per mm for Z set wildly wrong.



  • @leonmf For one @deckingman is right, the last one is definitely driven by the motor. You can even hear the high pitched sound of the motor running at a very high speed. Please post homez.g, homeall.g and maybe also your config.g.



  • @wilriker said in Duet Maestro Freefall:

    @leonmf For one @deckingman is right, ...................

    Manuel, mind if I correct your translation? That sounds a bit like "For once deckingman is right" To a native English speaker, that means the person is usually wrong but on this occasion was right. Now of course, you might have meant that I'm usually wrong, in which case no correction is needed 😉



  • @deckingman said in Duet Maestro Freefall:

    Manuel, mind if I correct your translation? That sounds a bit like "For once deckingman is right" To a native English speaker, that means the person is usually wrong but on this occasion was right. Now of course, you might have meant that I'm usually wrong, in which case no correction is needed 😉

    Thanks and I never mind being corrected! 🙂 Also I definitely did not want to say that you are usually wrong because I know better. 😂 (And I don't mean I know better not to say it but I know better that this would not be true).

    But one thing I wonder now: I wrote "For one" and wanted to say "First of all" whereas you corrected "For on_c_e" (which I would have known to have the meaning to described). So would my wording have expressed what I wanted to say?


  • administrators

    I agree, the motor is definitely driving the bed down.

    Is the printer definitely being controlled by a Duet Maestro? I believe M3D used other Duets in early machines, before the Maestro became available. So I am wondering whether your machine is running a Duet 2 Ethernet and the stall detection is configured to do something silly.



  • @wilriker said in Duet Maestro Freefall:

    But one thing I wonder now: I wrote "For one" and wanted to say "First of all" whereas you corrected "For on_c_e" (which I would have known to have the meaning to described). So would my wording have expressed what I wanted to say?

    Yes, you did say "for one" but it's not a common or normal expression. So the meaning of what you wrote was clear but it's an unusual thing to say. Hence it could easily be misinterpreted as a typo and the actual meaning was something else. Whilst I always enjoy these little chats, this probably isn't the time or place to discuss the finer points of English grammar. ☺



  • @deckingman said in Duet Maestro Freefall:

    Whilst I always enjoy these little chats, this probably isn't the time or place to discuss the finer points of English grammar. ☺

    You are right and we should put a stop to hijacking this thread. 😉 Thanks for clarification, though.



  • @dc42 This is definitely a Duet Maestro. I can post a board picture later if you want confirmation.

    Firmware Name: RepRapFirmware for Duet 2 Maestro
    Firmware Electronics: Duet Maestro 1.0
    Firmware Version: 2.01beta1(RTOS) (2018-06-23b1)
    Web Interface Version: 1.22.1

    Here is my current config:

    https://www.dropbox.com/s/aeppocjnft4zg4q/Promega Config.zip?dl=0

    I looked for an a M915 command in the files but found nothing, Typing M915 into the gcode console says not supported. Is there something else I should be looking at?



  • Could this by due to the fact that you have Deploy/retractprobe files in the sys directory there was a note on one of the firmware updates to remove them as they caused issues?



  • What confuses me more is that on the one hand the config defines

    M574 X2 Y2 Z2 S0 ; Set xy end-stops types (S0 is active low, applied to XY)
    

    a Z-max switch as an endstop in machine_endstoptypes.g and then also

    M574 X2 Y2 Z2 S0 ; Set xy end-stops types (S0 is active low, applied to XY)
    

    a switch-based Z-probe in machine-zprobe.g.

    Additionally at the start of homez.g there is

    ; Ignore Machine boundaries
    M564 H0 S0
    

    and I am not sure how this influences the latter

    ; Rapid Z until limit switch triggers
    G0 Z450 F1500 S1
    

    in the same file.


    So @LeonMF can you give some more information on what your endstop for Z actually is?



  • The way the config has been written with all those called macros is very confusing that's for sure it seems to be very over complicated


 

Looks like your connection to Duet3D was lost, please wait while we try to reconnect.