Magball Arms length and delta calibration



  • Hi, just a quick question on the actual length of Magball Arms and the value I should enter in M665 of config.g.
    This might be a silly question, but do I need to enter the length printed on the rod, or a sum of arm length and diameter ball couplers or something?
    Thanks!



  • If you purchased them from Haydn/ultibots etc, the number on the rod will be correct. The actual measurement is from the pivot point of both ends, which is the distance between the center of each ball-stud.



  • In other words, rods+ (radius of ballstud * 2)



  • Those are good starting points but the rod length you use might be longer or shorter depending on whether printed objects are correct in xy. Search forum there's a lot about this.



  • @DjDemonD:

    Those are good starting points but the rod length you use might be longer or shorter depending on whether printed objects are correct in xy. Search forum there's a lot about this.

    Thanks DJDemon, there is a lot infos indeed, some of which is rather confusing then helping. I am printing with ABS mostly, so a deviation of 0.1mm can also be the material shrinkage, etc. this was more a rithorik question so to say.

    I have never bothered to measure my rods actually, otherwise I would see, that the labeled length of 360.21mm is including the pivot point already. Again, a little silly, but the printer was busy as it poped into my mind…
    Previous build was coming from Think3DPrint3D with preconfigured FW, so I had no doubts there...

    Thanks everyone for help!



  • @Teilchen:

    @DjDemonD:

    Those are good starting points but the rod length you use might be longer or shorter depending on whether printed objects are correct in xy. Search forum there's a lot about this.

    Thanks DJDemon, there is a lot infos indeed, some of which is rather confusing then helping. I am printing with ABS mostly, so a deviation of 0.1mm can also be the material shrinkage, etc. this was more a rithorik question so to say.

    I have never bothered to measure my rods actually, otherwise I would see, that the labeled length of 360.21mm is including the pivot point already. Again, a little silly, but the printer was busy as it poped into my mind…
    Previous build was coming from Think3DPrint3D with preconfigured FW, so I had no doubts there...

    Thanks everyone for help!

    So use the 360.21 for your rod length in the FW and use 6 or 8 factor calibration in your bed.g NOT 7 or 9 factor.



  • Use the measured or given length then do 6 factor (or 8 if you want tilt compensation). If an object prints too big in x and y (should be the same amount too big in x and y otherwise you might have other issues) then make your rod length in M558 shorter then recalibrate and print again. Vice versa if object is too small. If the object is 1% too big make the rod length 1% shorter.



  • Thanks, but I recall David was suggesting to better use a M579 GCode to adjust the XY scale…



  • I tried using M579 on my Delta to fix a dimensional error on just one axis, but the prints came squeezed near the center (was using a negative Y value). Maybe this g-code is only meant for cartesian style printers. Do as Simon says and adjust the rod length for calibrating X-Y. Or try it, who knows, maybe it works for you.



  • Yeah I did not get far using scaling by that method, but do try it maybe its been tweaked for deltas, it would be a nice way to get correctly scaled parts, more direct, less trial and error, less plastic wasted.


  • administrators

    My recommendation is to print a grid pattern, then adjust the rod length to get straight lines and uniform spacing as far as possible. Then use M579 to correct any scaling.


Locked
 

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