Bed_Distance_sensor
Bed_Distance_sensor copied to clipboard
markniu
Probe goes nuts
This is my probe: BDsensorVer:V1.1 pandapi3d.com ,switch_mode=0,collision_homing=0,collision_cal=0
It is connected to the BL Touch port on a BTT Manta M5P [BDsensor] sda_pin: PC13 scl_pin: PC15 delay: 20 x_offset: -33 y_offset: 0 no_stop_probe: position_endstop: 0.5 collision_homing:0 collision_calibrate:0
KLIPPER and the bed sensor code were updated on Apr 16, the MCU was reflashed at that time. This has been going on since I installed the probe about a month ago.
It will work for a while then it says it is triggered when it is clearly not.
M119 x:open y:open z:open G28 Z triggered at 10.150 mm warning:triggered at 0mm, Please slow down the homing z speed Z triggered at 0.000 mm,auto adjusted. M119 x:open y:open z:open
The probe is actually about 120MM from the bed and the LED is OFF
I will then power cycle the printer and all is good for a while; then it will go nuts again
did you calibrate the bdsensor? and what's the data returned after M102 S-5 ?
Yes I ran M102 S-6, here is the M102 S-5 data: 1015 1015 1015 1015 1015 976 976 976 964 943 921 902 880 857 837 817 794 770 749 727 704 681 661 640 614 589 565 541 517 492 469 444 419 369 369 345 315 287 262
The probe is about 2mm from the tip of the nozzle, I tried making the gap smaller in order to get closer to 100 for the first reading but then I did not have 10 between the first and second values.
G1 Z1 BDSENSOR_DISTANCE 1.02mm
you can try to decrease the homing_retract_speed in stepper_z, for example homing_retract_speed:2 or disable the second homing by set the homing_retract_dist:0 in the stepper_z
and also attach klipply.log
As requested I added homing_retract_dist:0 in my configuration, a few minutes later I changed the filament in the printer and upon loading the new filament I had the error which I confirmed with: G28 Z triggered at 10.030 mm warning:triggered at 0mm, Please slow down the homing z speed Z triggered at -0.120 mm,auto adjusted.
The probe is 180mm from the bed and M119 correctly reports x:open y:open z:open
I am attaching klippy.log klippy-5.log
what's the returned value after M102 S-2 instead of M119, it seems the returned value is -0.12
I also suspect the electronic noise from other wires for example the motor or heater. could you try to test the homing by sending a lot of G28 Z to see if there is a problem while the E motor and heater are disabled? G28 Z G28 Z ....
G28 triggered at 10.150 mm warning:triggered at 0mm, Please slow down the homing z speed Z triggered at 0.000 mm,auto adjusted. M119 x:open y:open z:open M102 S-2 BDsensor:Out of measure Range or too close to the bed
Sadly M102 S-2 is useless in this case.
Further testing allowed me to get the error on demand, every time I loaded or unloaded the filament I could not home Z properly. Right now I am at the point where it seems that the probe cable does not like running in the cable chain alongside the extruder stepper motor cable (I have run many print jobs successfully so I don't think the hot end is involved, the extruder stepper motor gets to spin faster while loading and unloading the filament so that is my number one suspect). I have installed another cable on the outside of the cable chain and went from 100% failures to 0%. The sampling is still pretty small so I won't call it solved just yet but it looks much better.
Now I need to figure out a permanent way to affix the probe cable on the outside of the cable chain.
someone also report this electromagnetic interference recently, so now I just updated the new cable back to the same as first version bdsensor, that is shielded, twisted pairs and high purity copper wire for preventing EMI. it will be shipped with the bdsensor now instead of the old cables in the stock.
That sounds really nice…. Can I have one?
Le 28 avr. 2024 à 21:33, Mark @.***> a écrit :
someone also report this electromagnetic interference recently, so now I just updated the new cable back to the same as first version bdsensor, that is shielded, twisted pairs and high purity copper wire for preventing EMI. it will be shipped with the bdsensor now instead of the old cables in the stock. image.png (view on web) https://github.com/markniu/Bed_Distance_sensor/assets/13879340/1c1609a9-42f0-42f3-86dd-4f04e897a3d6 — Reply to this email directly, view it on GitHub https://github.com/markniu/Bed_Distance_sensor/issues/125#issuecomment-2081756850, or unsubscribe https://github.com/notifications/unsubscribe-auth/BICSRHUPXSW66G3LIZPB5TDY7WPP3AVCNFSM6AAAAABGZDSGM2VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDAOBRG42TMOBVGA. You are receiving this because you authored the thread.
That sounds really nice…. Can I have one?
yes, please send me your shipping address to my email: [email protected]
I am so sorry for that waste and thank you for the support and feedback those helpful information. the timer too close should be the new bug caused by the recent code committed that I have not test too much. and about the QGL there is a new quad_gantry_level_probe:1 that can let the probe up and down while doing QGL. sorry again.
Thank you so much for far more test , please send me your shipping address with email:[email protected], I would like to send you the 2 type of new sensor for test.
In January 2024, I also encountered issues where the timer to close error appeared with this sensor. Even with your code patched into Klippy, the error occurred intermittently when the sensor wasn't in use. This isn't a new bug, and I'm not the only
someone also report this electromagnetic interference recently, so now I just updated the new cable back to the same as first version bdsensor, that is shielded, twisted pairs and high purity copper wire for preventing EMI. it will be shipped with the bdsensor now instead of the old cables in the stock.
Hi @markniu, is it possible to buy these cables somewhere (preferably AliExpress)? And will the BD-Sensor M also be available at AliExpress in the near future?
So the new cable has been received and installed in the cable chain, no more problem with the probe getting lost in space. I really like this sensor, only problem I have is while running a Z_TILT_ADJUST the sensor will go to the left, probe the bed then move to the right and if the bed is skewed too much (looks like around 5mm) it just gives up along the line of: not triggered after full motion. My bed is 330 x 330 with the leadscrews 90mm away so it does not take that much to be off by 5mm.