Shifting BL Touch readings and Z Offset on SKR 1.4 Turbo PLEASE HELP

[NastyEwok]

I seem to be having very strange issues with my SKR 1.4 Turbo 4 x TMC2209 v1.2 drivers on UART.

I install the Firmware (Latest Marlin Bugfix, followed Gadget Angels Guide) reset the eeprom and return to defaults. Set the Z Offset with a 0.8 feeler gauge, and run BLTouch on a 7x7 pattern, 2 reads at each point.

First issue, My Z Offset seems to constantly be shifting, set it to -2.0, then check it again and it is -2.3, rehome a few times and check, start a job and just about scraping the bed, check again and it is back to -2.0, this has been going on for ages.

Have just ran 2 bed levels

back to back, bed heated to 55 and 2 completely different sets of results as below.

Have I got a faulty board here, or FW config issues...its driving me insane.

Can someone help please

0 1 2 3 4 5 6

0 -1.456 -1.369 -1.089 -0.919 -0.698 -0.515 -0.461

1 -1.331 -1.265 -1.014 -0.852 -0.626 -0.453 -0.410

2 -0.957 -0.914 -0.683 -0.556 -0.353 -0.202 -0.187

3 -0.753 -0.728 -0.512 -0.402 -0.208 -0.066 -0.059

4 -0.419 -0.424 -0.238 -0.154 +0.007 +0.118 +0.086

5 -0.160 -0.174 +0.013 +0.090 +0.263 +0.366 +0.330

6 +0.156 +0.120 +0.279 +0.336 +0.465 +0.520 +0.444

0 -0.747 -0.667 -0.384 -0.216 -0.003 +0.194 +0.236

1 -0.621 -0.556 -0.306 -0.142 +0.073 +0.260 +0.290

2 -0.242 -0.202 +0.028 +0.159 +0.352 +0.504 +0.521

3 -0.035 -0.015 +0.199 +0.322 +0.502 +0.654 +0.658

4 +0.297 +0.292 +0.479 +0.570 +0.715 +0.836 +0.802

5 +0.560 +0.540 +0.723 +0.814 +0.969 +1.081 +1.040

6 +0.879 +0.836 +0.993 +1.049 +1.166 +1.238 +1.157

What the heck is going on?

Marlin.zip

[radek8]

Uncomment:

define BLTOUCH_DELAY 500

define MONITOR_DRIVER_STATUS

define HYBRID_THRESHOLD

define TMC_DEBUG

What will Marlin answer for your M122 command?

I'm not sure about your value

define Z_MIN_PROBE_ENDSTOP_INVERTING

I think it should be set to "true"

[NastyEwok]

Marlin.zip

This is really getting silly now, No matter what I try I cannot get a first layer to go down at all.

I have now rebuilt my printer to make sure it was not something mechanical, and tried a different preconfigured, latest release 2.0.7.2 firmware from here.

https://oddity.oddineers.co.uk/2020/10/28/marlin-2-0-7-2-ender-3-pro-btt-skr-1-4-turbo-with-abl-using-bltouch-v3-1/

I have installed the firmware, Ran M502 and M500, then G28 to autohome, set the Z offset so a .8 feeler just fits under, run a 7x7 2 point bilinear mesh and then when I start the print it seems to have completely lost the bed level data.

Can anyone assist with this please, It is driving me insane

Here is the M122 report

Bilinear Leveling Grid: 0 1 2 3 4 5 6 0 -0.182 -0.102 -0.051 -0.093 +0.035 -0.030 +0.013 1 -0.249 -0.127 -0.052 -0.058 +0.099 +0.067 +0.143 2 -0.149 -0.068 -0.018 -0.063 +0.058 -0.013 +0.028 3 -0.230 -0.118 -0.048 -0.075 +0.069 +0.022 +0.088 4 -0.086 -0.014 +0.022 -0.049 +0.061 -0.028 +0.001 5 -0.140 -0.058 -0.007 -0.056 +0.064 -0.003 +0.046 6 +0.041 +0.075 +0.099 +0.007 +0.101 -0.008 +0.007 echo:busy: processing echo:Settings Stored (818 bytes; crc 2745) echo:Settings Stored (818 bytes; crc 2745) echo:busy: processing echo:busy: processing echo:busy: processing echo:busy: processing echo:busy: processing echo:busy: processing echo:busy: processing echo:Settings Stored (818 bytes; crc 53388) echo:Settings Stored (818 bytes; crc 54109) echo:enqueueing "M23 ce3pro~1.gco" echo:enqueueing "M24" echo:Now fresh file: ce3pro~1.gco File opened: ce3pro~1.gco Size: 34609 File selected echo:busy: processing echo:busy: processing echo:busy: processing echo:busy: processing echo:busy: processing echo:busy: processing

M122 SENDING:M122 axis:pwm_scale/curr_scale/mech_load|flags|warncount X Y Z E Address 0 0 0 0 Enabled false false false false Set current 600 600 600 500 RMS current 581 581 581 489 MAX current 819 819 819 689 Run current 18/31 18/31 18/31 15/31 Hold current 9/31 9/31 9/31 7/31 CS actual 9/31 9/31 9/31 7/31 PWM scale vsense 1=.18 1=.18 1=.18 1=.18 stealthChop true true true true msteps 32 32 32 32 tstep max max max max PWM thresh. [mm/s] OT prewarn false false false false triggered OTP false false false false pwm scale sum 29 29 33 11 pwm scale auto 1 1 0 0 pwm offset auto 91 90 0 44 pwm grad auto 28 26 0 27 off time 4 4 4 4 blank time 24 24 24 24 hysteresis -end 2 2 2 2 -start 1 1 1 1 Stallguard thrs 0 0 0 0 uStep count 84 556 140 516 DRVSTATUS X Y Z E sg_result 84 74 80 2 stst olb ola s2gb s2ga otpw ot 157C 150C 143C 120C s2vsa s2vsb Driver registers: X 0xC0:09:00:00 Y 0xC0:09:00:00 Z 0xC0:09:00:00 E 0xC0:07:00:00 Testing X connection... OK Testing Y connection... OK Testing Z connection... OK Testing E connection... OK X:29/9/-|TI| Y:29/9/-|TI| Z:33/9/-|TI| E:11/7/-|TI| X:29/9/-|TI| Y:29/9/-|TI| Z:33/9/-|TI| E:11/7/-|TI| X:29/9/-|TI| Y:29/9/-|TI| Z:33/9/-|TI| E:11/7/-|TI| X:29/9/-|TI| Y:29/9/-|TI| Z:33/9/-|TI| E:11/7/-|TI| X:29/9/-|TI| Y:29/9/-|TI| Z:33/9/-|TI| E:11/7/-|TI| X:29/9/-|TI| Y:29/9/-|TI| Z:33/9/-|TI| E:11/7/-|TI| X:29/9/-|TI| Y:29/9/-|TI| Z:33/9/-|TI| E:11/7/-|TI| X:29/9/-|TI| Y:29/9/-|TI| Z:33/9/-|TI| E:11/7/-|TI| X:29/9/-|TI| Y:29/9/-|TI| Z:33/9/-|TI| E:11/7/-|TI| X:29/9/-|TI| Y:29/9/-|TI| Z:33/9/-|TI| E:11/7/-|TI| X:29/9/-|TI| Y:29/9/-|TI| Z:33/9/-|TI| E:11/7/-|TI| X:29/9/-|TI| Y:29/9/-|TI| Z:33/9/-|TI| E:11/7/-|TI| X:29/9/-|TI| Y:29/9/-|TI| Z:33/9/-|TI| E:11/7/-|TI| X:29/9/-|TI| Y:29/9/-|TI| Z:33/9/-|TI| E:11/7/-|TI| X:29/9/-|TI| Y:29/9/-|TI| Z:33/9/-|TI| E:11/7/-|TI| X:29/9/-|TI| Y:29/9/-|TI| Z:33/9/-|TI| E:11/7/-|TI| X:29/9/-|TI| Y:29/9/-|TI| Z:33/9/-|TI| E:11/7/-|TI| X:29/9/-|TI| Y:29/9/-|TI| Z:33/9/-|TI| E:11/7/-|TI| X:29/9/-|TI| Y:29/9/-|TI| Z:33/9/-|TI| E:11/7/-|TI| X:29/9/-|TI| Y:29/9/-|TI| Z:33/9/-|TI| E:11/7/-|TI| X:29/9/-|TI| Y:29/9/-|TI| Z:33/9/-|TI| E:11/7/-|TI| X:29/9/-|TI| Y:29/9/-|TI| Z:33/9/-|TI| E:11/7/-|TI| X:29/9/-|TI| Y:29/9/-|TI| Z:33/9/-|TI| E:11/7/-|TI| X:29/9/-|TI| Y:29/9/-|TI| Z:33/9/-|TI| E:11/7/-|TI| X:29/9/-|TI| Y:29/9/-|TI| Z:33/9/-|TI| E:11/7/-|TI| X:29/9/-|TI| Y:29/9/-|TI| Z:33/9/-|TI| E:11/7/-|TI| X:29/9/-|TI| Y:29/9/-|TI| Z:33/9/-|TI| E:11/7/-|TI| X:29/9/-|TI| Y:29/9/-|TI| Z:33/9/-|TI| E:11/7/-|TI| X:29/9/-|TI| Y:29/9/-|TI| Z:33/9/-|TI| E:11/7/-|TI| X:29/9/-|TI| Y:29/9/-|TI| Z:33/9/-|TI| E:11/7/-|TI| X:29/9/-|TI| Y:29/9/-|TI| Z:33/9/-|TI| E:11/7/-|TI| X:29/9/-|TI| Y:29/9/-|TI| Z:33/9/-|TI| E:11/7/-|TI| X:29/9/-|TI| Y:29/9/-|TI| Z:33/9/-|TI| E:11/7/-|TI|

[christian-hellwig]

Got the very same issue. But in my case with an capacitiv sensor. Ran a g28 and the a g29 approx all 20 minutes. After 90 minutes the level measurements went from -0.012mm to 6.423mm. Needless to say thats not a usable situation.

[christian-hellwig]

Okay I managed to find a solution for my issue. Still got a little temp drift, but that maybe my sensor. I had a problem with my endstop connection. How did you connect the blt touch and what is your endstop configuration, i mean in hardware

[zed-maker]

Have you tried setting your initial z offset gap to less than 0.8mm? That seems too big to get bed adhesion. I’ve always used 0.1 to 0.15mm. I would be surprised if you get any adhesion at 0.8mm. Unless you mean 0.08mm that is. I found bed adhesion to be hugely improved when I upgraded from Marlin 1.1.9 to 2.0bugfix . Variations in BL Touch readings that I have come across were from the mounting nuts and bolts on the BL touch to the hot end working themselves loose. Check all your screws, bolts and mechanicals on the entire printer to make sure they are all still tight. One loose bolt is all it takes to stuff things up. Good luck and I hope you work it out.