1.18 Z height offset question



  • David and all, I upgraded to 1.18 release this morning from 1.17 on one of my deltas. This printer has FSR/JohnSL probing and calibrates to a deviation of 0.01 and I always got a perfect first layer. I saw in the release notes the comment about Z offset:

    You may find that your Z probe trigger height is slightly higher than before, so you should re-measure it

    With the new baby stepping feature I thought it made sense to proceed and print a single layer object that I use to tune settings and use baby stepping to adjust Z. I did this and had to lower Z by 0.20mm. This seems like a a pretty big difference, what are folks experience with Z probe trigger height in 1.18?

    Now the odd thing and question. I subtracted the extra offset from my G31 Z to get "G31 Z-0.35" in config.g. I deleted config-override.g and restarted and then ran delta auto calibration and committed with M500 (again with a 0.01 deviation). I then attempted to print the single layer object and was surprised to find I needed to baby step down 0.20mm again! I verified that my G31 Z-.35 is in config.g and now in config-override,g. Any idea what's going on?

    cheers,
    Michael


  • administrators

    Was your G31 Z offset previously -0.15 or -0.55 ?

    0.2mm is a larger change than I would expect.



  • It was previously -.15 mm


  • administrators

    I think that reduction in trigger height was spurious, nothing to do with the firmware upgrade. If anything, the trigger height should be higher using 1.18, not lower.



  • I've gone through the printer mechanics and configs and it is still behaving as described. Multiple deletes of config-override and resets. It's almost as if G31 Z offset is not being applied.



  • David, I downgraded to 1.17 and my original Z offset is working perfectly. I'll re-convert to 1.18 when I get time early this week and try again.


  • administrators

    Michael, have you tried printing using 1.18 and your old height offset?


  • administrators

    @mhackney:

    I've gone through the printer mechanics and configs and it is still behaving as described. Multiple deletes of config-override and resets. It's almost as if G31 Z offset is not being applied.

    That can happen if you put the G31 command earlier in config.g than the M558 command. This is because RRF stores different G31 parameters for different types of Z probe, to allow you to use more than one probe.



  • @dc42:

    Michael, have you tried printing using 1.18 and your old height offset?

    David, yes that is actually what I was trying to articulate above. With the same height offset I had to baby step down an additional 0.20mm. If I then change my height offset to -.35mm I STILL have to baby step down .20mm. That's what's strange. I see your second post about order so let me verify that, although once committed to config-override.g the correct order is captured, correct?



  • @dc42:

    @mhackney:

    I've gone through the printer mechanics and configs and it is still behaving as described. Multiple deletes of config-override and resets. It's almost as if G31 Z offset is not being applied.

    That can happen if you put the G31 command earlier in config.g than the M558 command. This is because RRF stores different G31 parameters for different types of Z probe, to allow you to use more than one probe.

    Understood and I do have them in the correct order. I actually take advantage of the ability to support different probes on one of my test machines that has the IR probe, FSRs and an accelerometer probe on it so I can do comparisons. On this machine I have:

    [[language]]
    ; Z probe and compensation definition
    M558 P4 X0 Y0 Z0 F250 H3 I1 T3500
    
    ; grid compensation
    M557 R110 S15
    
    G31 T4 X0 Y0 Z-.20			; Set the z probe offset and threshold
    
    

  • administrators

    As long as you had the correct probe type selected when you sent the G31 command, then M500 will save it correctly in config-override.g. The G31 commands in that file use the T parameter to say what probe type they refer to. As far as I know, that parameter works correctly. If you suspect it doesn't, run G31 Tn where n is your probe type (as in M558 P parameter) and it will tell you the trigger height that it is using.



  • Today I built a new CF card from a backup set of config files I had. I turned off the printer, installed the new card and restarted. Verified that there was no config-override.g, checked my offsets, Z height and other calibration settings. All looked good. Calibrated, looked good (0.01 deviation) then I verified height and it seemed spot on. Started a print and the first layer looks fine. I'm waiting for it (1 layer) to complete so I can verify thickness.

    But it does seem like I must have displeased the 3D Printing Deities this weekend and now I'm back in good standing.



  • hopefully my red head sacrifice takes hold and the gods will be pleased with me. we'll find out tomorrow, I don't wish to pursue any 3d printing tonight.


Log in to reply