JohnSL FSR Sensors and Bed Printing Too Close at Towers



  • I've been trying to tune my JohnSL sensors and it seems to consistently place my nozzle too close to the bed when it prints under each tower - no matter what I alter, it seems to follow that consistent pattern (closer to the bed near Z,Y,X and then moves farther away between towers.

    I can print objects around ~125mm in diameter and they come out good, but as the towers are approached the first layer varies (being consistently thin as the nozzle approaches the tower).

    Using a sheet of paper, it doesn't seem like there's much if any difference (maybe -0.02 to -0.04 at most) between probe points and the middle.

    I may try to run another paper check today if I have time.

    My Firmware Version: 1.18.1 (2017-04-09)

    What's a good checklist to go through to make sure I'm not missing something obvious?

    Here's what I've checked:

    Belts (fairly taut, no slipping)
    Steps per mm (80)
    Frame is square (or rather triangular)
    Manually checked Delta Radius (I don't think it is this because the shape is not completely concave or convex, only alters near towers)
    My endstops usually read around -0.20, -0.20, 0.40 for ZYX respectively - maybe vary 0.05 between auto-levelings from different settings
    I've altered the pressure setting for G31, but that doesn't seem to change the tower closeness
    My delta radius usually comes out to around 142 to 142.5
    I've done several variations of different probe points and probe radiuses but the tower closeness appears repeatable
    Swapped around bed clips and loosened and then removed screws to allow Onyx heat bed to float under frame
    I think Bed tilt from auto-leveling is under half a degree in X and Y
    Added grid leveling after G31, but it doesn't seem to be any worse or any improvement



  • What is your M557 Rxxx set to?



  • I assume your FSR's are mid way between towers like mine? If so, the bed deflects more due to no support at the towers which lifts the bed every so slightly on the opposite side using 2 FSR's as a pivot point. I went to probing 3 sets of 3 points all in a line from my FSR's to the bed center and don't have that problem anymore. Here are my points:

    G30 P0 X0.00 Y-100.00 Z-99999 H0            ; between X-Y towers
    G30 P1 X86.60 Y50.00 Z-99999 H0                             ; between Y-Z towers
    G30 P2 X-86.60 Y50.00 Z-99999 H0            ; between Z-X towers
    G30 P3 X0.00 Y-75.00 Z-99999 H0                             ; between X-Y towers
    G30 P4 X64.95 Y37.5 Z-99999 H0                ; between Y-Z towers
    G30 P5 X-64.95 Y37.5 Z-99999 H0                             ; between Z-X towers
    G30 P6 X0.00 Y-50.00 Z-99999 H0                             ; between X-Y towers
    G30 P7 X43.30 Y25.00 Z-99999 H0                             ; between Y-Z towers
    G30 P8 X-43.30 Y25.00 Z-99999 H0            ; between Z-X towers
    G30 P9 X0 Y0 Z-99999 H0 S6                        ; center and auto-calibrate 6 factors



  • @Alexander:

    I assume your FSR's are mid way between towers like mine? If so, the bed deflects more due to no support at the towers which lifts the bed every so slightly on the opposite side using 2 FSR's as a pivot point. I went to probing 3 sets of 3 points all in a line from my FSR's to the bed center and don't have that problem anymore. Here are my points:

    G30 P0 X0.00 Y-100.00 Z-99999 H0            ; between X-Y towers
    G30 P1 X86.60 Y50.00 Z-99999 H0                             ; between Y-Z towers
    G30 P2 X-86.60 Y50.00 Z-99999 H0            ; between Z-X towers
    G30 P3 X0.00 Y-75.00 Z-99999 H0                             ; between X-Y towers
    G30 P4 X64.95 Y37.5 Z-99999 H0                ; between Y-Z towers
    G30 P5 X-64.95 Y37.5 Z-99999 H0                             ; between Z-X towers
    G30 P6 X0.00 Y-50.00 Z-99999 H0                             ; between X-Y towers
    G30 P7 X43.30 Y25.00 Z-99999 H0                             ; between Y-Z towers
    G30 P8 X-43.30 Y25.00 Z-99999 H0            ; between Z-X towers
    G30 P9 X0 Y0 Z-99999 H0 S6                        ; center and auto-calibrate 6 factors

    I'll try this out with both between tower and only tower versions and see what it looks like.

    It seems like some sort of deflection could be the issue here. I'm thinking I may try some silicone tape that's much thinner and let my bed rest on that instead of the thicker silicone pads to see what happens.



  • @jmjcoke:

    What is your M557 Rxxx set to?

    I had around a 100-120 radius for that, but I quit using M557 to focus on the G31 calibration first - since I think something with it was causing the issue.



  • How are the FSR sensors working for you? Have you compared it to the DC42 IR sensor?


Locked
 

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