probing is way out.. well kinda maybe?

  • ok, i think this is the final issue that i have with my robo swap.

    i finally have all of the probing movements down, i ended up moving my machine to a magnetic build plate, (buildtak), i am getting way more consistent probes now, only my deviations are atrocious. almost half a mm out on the highest deviation which makes no sense.


    so i have greatly dug into this and i believe that its likely that my bed is warped beyond use. hopefully some one can convince me other wise...

    so i initially ran a g32, which came up with the above height map. i didnt think that it could possibly be that bad so i automatically assumed it was the probe that was messing up.

    i decided to view the individual probes with a little macro i wrote, and averaged out the probes to get a rough guestimate on what i was dealing with... wow was that an eye opener.

    pre shim probe.JPG

    firstly my glass bed is held down with magnets that are attached to the Y axis gantry. i assumed that each of the magnets would have been put in with some sort of precision... uhhh not at all

    bed shimming.JPG

    so while i wait for my actual metal shim sheets to come in, i have just used varying sheets of paper as shims and tried to get them as close as i could to the highest magnet.

    after i had everything shimmed up to my satisfaction i re-ran the probing macro i made and it gave me this.

    Post shim probe.JPG

    as you can see, the shimming did make a difference but it just tells me that the bed is gone whack yo.

    can anyone confirm if these should be within the specs of a typical IR sensor on a buildtak build plate? perhaps @dc42 (keep in mind i believe this is a clone of yours)

    This also brings up another valid point, what should i replace the bed with?
    ive already got a 6.5mm thick plate of aluminum, and a 1.5mm thick plate of aluminum. im also considering redesigning my y axis gantry to allow me to put screws and springs so i can manually level it.

    any ideas would be appreciated.


  • Had similar problems with my IR sensor and buildtak. Turns out - nothing wrong with my bed, i just couldnt make IR sensor work with it. If you have even a little bit of plastic melted on buildtak, then it doesnt work properly. And my main problem was ambient lighting, affects alot. So i switched to BLtouch and no any problem with probe anymore at all.

  • well this buildtak is brand new, not even a print on it. so it seems a little weird that its still messed up.

    how can a guy manually level it, and just ignore the zprobe with the duet?

  • @aidar ive already ripped through all of that,

    at the end of the day, the zprobe ir differential just isnt a sensitive enough probe on my machine.

    manually leveling the bed was the way to go

    instead of it being 1.5mm off the whole bed, its only actually 0.25 deviation across the whole bed off a 0.2 feeler gauge.

    from this

    to this
    manual probe.JPG .

    thanks for trying to sort me out though i really appreciate it!

Log in to reply