Bed leveling options



  • Of these options, which direction should I head for bed leveling? I print on glass so inductive is out and capacitive is likely out due to the temperature sensivity and where my printer is (table is directly above heating/cooling vent).

    BL-touch/etc
    Piezo
    IR
    Wait for Duet to support stallguard

    My printer is a corexy custom build with Titan Aero extruder and I do have plenty of room for a sensor so that's not a factor.

    I tend to want to go with the 'best' solution going forward, not just what's commonly used now.


  • administrators

    BLTouch in theory works on any surface, but some users find it doesn't give very reproducible results. Piezo should also work on any surface, but you have to probe with the nozzle hot enough to melt any filament that is stuck to the nozzle. IR works well on glass if either the glass is black or you put a black surface under it, and you don't put a coating on the glass (this is what I use on my Cartesian printer). If you coat the glass with glue, hairspray etc. then it gives less reproducible results, at least on transparent glass.

    HTH David



  • Those are the exact thoughts I've had on those 3, each with it's own drawback. I do use glue on glass.

    So, I notice the lack of comment on Duet/stallguard? I truly hope that's an option coming down the pipeline because I can wait and continue with manual leveling for now. Otherwise, piezo may be my direction.


  • administrators

    Using stallguard on the Z axis to do bed probing may not be a good option. The resolution of stallguard is either 1 or 4 full steps depending on how it is configured. So in order to achieve 0.01mm accuracy, at x16 microstepping you would need your Z steps/mm to be at least 1600 and preferably 6400. With such a high steps/mm, the potential nozzle contact force exerted by the Z axis would be very high, and your build would have to be very strong to resist it without damage.

    You would also need to have a single stepper motor driving 3 leadscrews. The 3 leadscrews are so that bed doesn't have significantly different amounts of give at different probe points, and the single motor would be needed to avoid getting out of sync when the stall occurs.

    But if you want to try it, you can do some experiments even before the firmware has stallguard support. See if you can find a Z motor current that will drive the head down into the bed but cause no damage when it stalls. Then check that you not getting any flexing of the bed, wherever on it you probe.



  • Since I switch bed surfaces all the time (most of them are 4mm float glass, but also PC and PEI) AND I use 3Dlack, I have a BLTouch installed since I wanted to physically probe the bed. I tried the IR probe, but with the different kind of surfaces I use, it wasn't working well in all cases. On clear glass with the ALU bed paitned black, the IR probe was very accurate.
    I can't complain about accuracy for the BLTouch.
    But I don't print at 0,05mm, seldom at 0.1 and most of the time it's in the range 0,2 to 0,5mm …



  • I can see how all that would be problematic. I run a single motor on dual Z screws that are 2mm which would not be ideal for using stall.



  • Well I obviously have an interest in this but what everyone who uses piezo says after they get it working, (and it may take a little tweaking and tuning to get there) is "wow, this thing is amazing, my first layers are perfect now" or words to that effect.

    So you know where I would go. The only other z-sensing system that is comparable in terms of convenience, accuracy, repeatability and versatility (use any surface) is the Smart Effector which works in the same way (but using a different sensing modality) and this is currently aimed at deltas although I'm sure you could adapt it with clever mounting, you can definitely adapt piezo to almost any printer.


 

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