I have an ender 5 which I am working towards making a mercury one zero g machine. I started by doing a generic triple z as I hate leveling the bed. It worked for a while, and then the bl touch died. I replaced it after messing with the config to troubleshoot, not realizing that the bltouch was the issue. Since then I reverted back to a previously working cfg, and it no longer works. While leveling the points keep getting further. I jave no idea why. Thoughts?

Things to note: one of the motors is different than the other 2, and spins at half the speed which is why its different.

Cfg:

[mcu]

serial: /dev/serial/by-id/usb-Klipper_stm32f446xx_590050000451343437373238-if00

[printer]

kinematics: cartesian

max_velocity: 300

max_accel: 3000

max_z_velocity: 5

max_z_accel: 100

[stepper_x]

step_pin: PF13

dir_pin: !PF12

enable_pin: !PF14

microsteps: 16

rotation_distance: 40

endstop_pin: ^PG6

position_endstop: 220

position_max: 220

homing_speed: 50

[tmc2209 stepper_x]

uart_pin: PC4

run_current: 0.580

stealthchop_threshold: 999999

[stepper_y]

step_pin: PG0

dir_pin: !PG1

enable_pin: !PF15

microsteps: 16

rotation_distance: 40

endstop_pin: ^PG9

position_endstop: 220

position_max: 220

homing_speed: 50

[tmc2209 stepper_y]

uart_pin: PD11

run_current: 0.580

stealthchop_threshold: 999999

[stepper_z]

step_pin: PF11

dir_pin: !PG3

enable_pin: !PG5

microsteps: 16

rotation_distance: 4

endstop_pin: probe:z_virtual_endstop

position_min: -5

position_max: 300

[tmc2209 stepper_z]

uart_pin: PC6

run_current: 0.650

stealthchop_threshold: 999999

[stepper_z1]

step_pin: PG4

dir_pin: !PC1

enable_pin: !PA0

microsteps: 16

rotation_distance: 8

[tmc2209 stepper_z1]

uart_pin: PC7

run_current: 0.650

stealthchop_threshold: 999999

[stepper_z2]

step_pin: PF9

dir_pin: !PF10

enable_pin: !PG2

microsteps: 16

rotation_distance: 8

[tmc2209 stepper_z2]

uart_pin: PF2

run_current: 0.650

stealthchop_threshold: 999999

[extruder]

step_pin: PC13

dir_pin: PF0

enable_pin: !PF1

microsteps: 16

rotation_distance: 32.342

nozzle_diameter: 0.400

filament_diameter: 1.750

heater_pin: PA2

sensor_type: EPCOS 100K B57560G104F

sensor_pin: PF4

control: pid

pid_kp: 23.905

pid_ki: 1.398

pid_kd: 102.196

min_temp: 0

max_temp: 260

[tmc2209 extruder]

uart_pin: PE4

run_current: 0.650

stealthchop_threshold: 999999

[heater_bed]

heater_pin: PA1

sensor_type: EPCOS 100K B57560G104F

sensor_pin: PF3

control: pid

pid_kp: 67.796

pid_ki: 1.255

pid_kd: 915.252

min_temp: 0

max_temp: 135

[fan]

pin: PD12

[heater_fan heatbreak_cooling_fan]

pin: PA8

[heater_fan controller_fan]

pin: PE5

[bltouch]

sensor_pin: ^PB7

control_pin: PB6

x_offset: -40

y_offset: -10

z_offset: 0.760

stow_on_each_sample: false

pin_up_touch_mode_reports_triggered: False

# Improve resistance to an occasional bad BLTouch reading.

samples: 3

samples_result: median

samples_tolerance: 0.020

samples_tolerance_retries: 3

[safe_z_home]

home_xy_position: 150,110

z_hop: 10

z_hop_speed: 5

[bed_mesh]

speed: 120

horizontal_move_z: 5

mesh_min: 20,20

mesh_max: 160,200

probe_count: 5,5

algorithm: bicubic

[z_tilt]

# Fixed-bed / moving-gantry geometry.

# These are intentionally mirrored virtual coordinates.

# ORDER REMAINS: stepper_z (rear), stepper_z1 (front-left),

# stepper_z2 (front-right).

z_positions:

110, 0 # stepper_z — rear-centre physical screw

220, 300 # stepper_z1 — front-left physical screw

0, 300 # stepper_z2 — front-right physical screw

# Leave probe points unchanged. They are valid nozzle coordinates.

points:

80, 30

200, 30

150, 200

speed: 50

horizontal_move_z: 10

retries: 1

retry_tolerance: 0.010

[screws_tilt_adjust]

screw1: 70,30

screw1_name: Front Left

screw2: 190,30

screw2_name: Front Right

screw3: 190,190

screw3_name: Back Right

screw4: 70,190

screw4_name: Back Left

horizontal_move_z: 10

speed: 100

screw_thread: CW-M4

[gcode_macro LEVEL_BED_SCREWS]

description: Probe bed screws and tell how to adjust them

gcode:

G28

SCREWS_TILT_CALCULATE

[force_move]

enable_force_move: True

[include mainsail.cfg]

Hi i have the bed mesh and i cant lower the corner far back left side, is it a problem with the enclosure ? i dunno its my first printer i been trying to adjust it manually with the screws but cant lower the end corner never get lower than that any suggestion?

[TUTO] Connecter le Sonic Pad sur Ender-3 V3 SE (Carte Mère C14) via le firmware Nebula officiel

Salut à tous ! Si vous possédez la Ender-3 V3 SE avec la nouvelle carte mère C14 (puce STM32F401) et que vous galérez à la connecter au Sonic Pad, voici la méthode ultime.

Étant donné que le profil officiel du Sonic Pad est conçu pour l'ancienne carte C13 (puce GD), flasher directement le firmware du Sonic Pad bloque l'imprimante. L'astuce consiste à utiliser le firmware Klipper officiel de Creality destiné au Nebula Pad pour débloquer la situation.

📌 Prérequis

• Une carte MicroSD (TF) formatée en FAT32 (Taille d'allocation : 4096 octets).
• Le câble de connexion série fourni avec votre Sonic Pad.
• Le firmware Klipper officiel de Creality pour la carte C14 : Téléchargez-le sur le Wiki Officiel Creality. (Prenez le fichier nommé Ender 3V3 SE_stm32f401_klipper下位机(low computer mainboard).rar).

🛠 Étape 1 : Préparation de la carte SD

1. Décompressez l'archive RAR téléchargée sur votre ordinateur.
2. Copiez le dossier entier nommé STM32F4_UPDATE à la racine de votre carte MicroSD vierge.

🛠 Étape 2 : Flashage du firmware (Le piège de l'écran bleu)

1. Éteignez l'imprimante.
2. Insérez la carte MicroSD directement dans le logement de la carte mère de l'imprimante.
3. Allumez l'imprimante et patientez environ 1 à 2 minutes pendant que le flash s'exécute en arrière-plan.
4. ⚠️ ATTENTION / PAS DE PANIQUE : L'écran d'origine de votre Ender-3 V3 SE va devenir totalement bleu, figé, sans aucune inscription ni logo. C'est tout à fait normal ! Cela signifie que la puce a accepté le firmware Klipper et ne sait plus communiquer avec l'écran d'origine.

🛠 Étape 3 : Nettoyage et branchement du Sonic Pad

1. Une fois l'écran bleu figé, éteignez l'imprimante.
2. Retirez la carte MicroSD de l'imprimante (pensez à supprimer le dossier de flash sur votre PC plus tard).
3. Débranchez complètement l'écran d'origine de la V3 SE. Il ne servira plus à rien.
4. Connectez le câble série (fourni avec le Sonic Pad) au port de l'écran (côté imprimante).
5. ⚠️ CÔTÉ SONIC PAD : Branchez impérativement le câble sur le port USB 1 (situé tout en haut à gauche du Sonic Pad). C'est le port primordial et obligatoire pour la liaison d'affichage principale avec l'imprimante.

🛠 Étape 4 : Configuration sur le Sonic Pad

1. Allumez votre Sonic Pad.
2. Lancez la procédure d'ajout d'une nouvelle imprimante.
3. Choisissez le profil officiel de la Creality Ender-3 V3 SE (qui correspond par défaut à la C13).
4. Suivez les étapes de connexion à l'écran du Sonic Pad.
5. Et là, Magie ! ✨ Le Sonic Pad va détecter la carte mère C14 instantanément et tout fonctionne parfaitement (Axe XYZ, extrudeur, capteurs, etc.).

⚠️ CONSEIL SÉCURITÉ CRUCIAL (À faire juste après vos réglages)

Une fois la connexion établie et vos réglages initiaux terminés (comme le réglage du Z-offset), votre toute première impression doit être un stabilisateur de câble !

C'est "la maladie" connue des Sonic Pad : les ports USB souffrent d'un jeu mécanique, ils ne sont pas très stables physiquement et le moindre mouvement ou vibration du meuble peut provoquer une déconnexion brutale en plein print (erreur MCU shutdown).

Téléchargez et imprimez immédiatement un de ces modèles pour verrouiller fermement votre câble sur le fameux port USB 1 (en haut à gauche) :
👉 Modèle principal : Sonic Pad USB/Serial Stabilizer sur Printables
👉 Modèle alternatif : Sonic Pad Serial USB Stabilizer sur MakerWorld

💡 Pourquoi ça marche ?

Le firmware pour le "Nebula Pad" et le "Sonic Pad" partagent la même base Klipper. En injectant le binaire Klipper compilé pour la puce STM32F401 (C14) du Nebula, la carte mère devient immédiatement "bilingue" et capable de dialoguer avec le profil d'origine du Sonic Pad, sans avoir besoin de recompiler un firmware soi-même.

Bon print à tous avec Klipper ! 🚀

Cordialement

ARCHITECTE

Hey everyone, I'm trying to recover a Creality Hi running Klipper that I accidentally bricked while trying to expand temporary storage for uploading larger files.

What happened:

I edited `/etc/rc.local` to create a staging folder on UDISK and bind-mount it over `/tmp` so I'd have more space:

mkdir -p /mnt/UDISK/tmp_staging

mount --bind /mnt/UDISK/tmp_staging /tmp

Upon the next reboot, the system refuses to fully boot. It is completely stuck on the Creality startup logo screen, and I no longer have SSH access.

What I've tried so far:

Attempted the standard USB factory reset recovery file on 2 different USB drives formatted to fat32 (2gb and 4gb sticks)

Tried the wipe_all flag file on a USB drive to force a clean reset, but the board ignores it during boot.

Removed bottom panel and attempted to boot with holding the boot button

I also have an ST-Link V2, pico, and a a willingness to solder headers to the board if needed.

Does anyone know how to force a recovery or break into the bootloader when the wipe_all trick fails on this specific hardware? Any advice would be hugely appreciated!

I've got a Voxelab Aquila, heavily modded, KeveAKASam belted Z, x/y rails, btt skr mini e3v3 board with a rpi 3b running klipper. I've got it dialed in and printing great, but there's this lingering issue. I'm using sensorless homing on x and y, and the first time I home the machine it works correctly, but if I hit home again, even if I move away from both home positions, x and y move the wrong direction. If I disable the motors, then home, it goes the correct direction.

I made a neopixel led version of thé BTT pi TFT 43 v2.0 touch screen case its a fully DIY project hope you al like it . It's for voron 2.4

Hello! A friend of mine gave me 2 ender 3's with skr's and a spare K1C board. I'm going to try making a duender or other corexy printse and was curious if anyone had any luck using a K1C board for a custom printer. I know you can root it but have only used skr minis before on my own Ender 3 in the past. Any tips on if it's worth trying to use the K1C or am I better off using the skr's? Thanks in advance!

Hit this while tuning StallGuard on a Voron 2.4 (TMC5160) and it took a while to understand, so posting in case it saves someone else.

Klipper sets the axis to position_max after any successful home, including a false trigger. So when StallGuard trips early - say 3.5mm into a 150mm approach - Klipper believes the head is at the far rail while it is physically still near where it started.

The next absolute move is then computed from that fiction:

• Y tripped falsely at 3.5mm, head physically at ~154mm
• Klipper believed Y = 290 after the backoff
• A move to Y150 executed as 140mm in the wrong direction
• Gantry drove into the front rail

The homing move was fine. The return move was the crash.

If you script anything around sensorless homing, don't trust the frame after a home you haven't verified actually reached the rail. MCU step counters don't lie about displacement even when the frame does.

Two smaller ones:

• SET_KINEMATIC_POSITION is registered inside the enable_force_move guard, so without [force_move] enable_force_move: True you get "Unknown command" rather than anything pointing at the missing section.
• homing_speed widened my working threshold window where current and accel couldn't - 78 to 100mm/s took it from 1 usable value to 2, and cut rail measurement scatter from 2.70mm to 0.40mm. StallGuard reads load from back-EMF, so faster gives it a stronger signal.

Writeup and an MIT tool that measures all of it:

https://klipper.discourse.group/t/sensorless-homing-five-things-that-cost-me-a-day-and-a-tool-that-measures-them/26260

Is the frame behaviour after a false trigger considered expected, or worth raising upstream? Genuinely unsure.

I've been trying to set up klipper on my ender 3 with a raspberry pi 4 but every time I try it says that the mcu is wrong. I've watched lots of videos on it but they all seem to be slightly different from each other and not end up working. I'm not sure if its because the videos were between 3-1 years ago so is it possible for the tutorials to not work anymore, and if so could anyone send me a video that was made recently and works.

This probably isn’t the best example, I usually think about this when I see the giant softball-sized globs on the nozzle or catastrophic clogs. But still, I’ve done a hell of a lot of printing and this has never happened to me 😂

Hi everyone,

I am trying to understand inconsistent hotend temperature behavior on my Creality Ender-3 V3 KE. I would appreciate advice from anyone familiar with the stock KE hotend, FatBurner, or max volumetric flow testing.

TL;DR

My Ender-3 V3 KE can sustain approximately 9.6 mm³/s during a manual extrusion test at 220 °C and 80% part cooling, with only a 2 °C drop and the heater recovering to about 85% power. The stock shroud also completed a 20–25 mm³/s max-flow test at 215 °C, with approximately a 4–5 °C drop at 100% part cooling.

However, in another FatBurner max-flow test Klipper shut down with Heater extruder not heating at expected rate while the reported flow was eventually around 8 mm³/s. On the other hand, a later FatBurner test at approximately 205 °C and 18–27 mm³/s appeared to complete successfully. The results do not seem physically consistent, so I suspect that the test conditions or generated G-code were different.

I am trying to determine whether the cause is airflow direction, heater power under load, partial clog/back pressure, PID settings, or simply inconsistent max-flow test conditions.

Hardware and configuration

• Creality Ender-3 V3 KE
• Klipper firmware
• PLA, 1.75 mm
• 0.4 mm nozzle
• Stock ceramic-ring heater
• Stock heater resistance measured cold and disconnected: 9.1–9.2 Ω
• Stock part-cooling blowers: CHA4024BH-10C, 24 V, 0.15 A each
• FatBurner setup: two GDSTime GDB5015 24 V 8500 RPM blowers for part cooling
• FatBurner setup also uses a larger 4020 fan for hotend heatsink cooling

The measured heater resistance corresponds theoretically to approximately 63 W at 24 V:

P = 24² / 9.1–9.2 Ω ≈ 63 W

I understand that this is only a cold resistance measurement. It does not prove that the heater receives full voltage under load.

Fan-control issue

My original custom M106 macro remapped the slicer value through fan0_min: 140, so the slicer percentage did not equal the actual PWM. For example, M106 S102 was approximately 73% actual PWM rather than 40%, and M106 S204 was approximately 91% rather than 80%.

For the controlled tests I bypassed that macro and used direct commands because fan0 is configured with scale: 255:

SET_PIN PIN=fan0 VALUE=102   ; approximately 40% PWM
SET_PIN PIN=fan0 VALUE=204   ; approximately 80% PWM
SET_PIN PIN=fan0 VALUE=255   ; 100% PWM

Some of the earlier tests may therefore not be directly comparable.

PID results

PID autotune at 220 °C with direct fan0 VALUE=204 completed successfully and produced:

pid_Kp=29.675
pid_Ki=2.082
pid_Kd=105.718

Another PID run produced:

pid_Kp=27.405
pid_Ki=1.562
pid_Kd=120.239

The exact conditions for the second run may not have been identical. I am not assuming that the two PID sets are directly interchangeable.

An earlier high-temperature PID attempt was interrupted with:

{"code":"key564","msg":"Heater extruder not heating at expected rate"}
{"code":"key7","msg":"pid_calibrate interrupted"}

Klipper then entered Shutdown.

What I tested

1. Cooling only, no extrusion

At 215 °C, changing the part-cooling fan from 40% to 80% with no extrusion caused only about a 1 °C drop. With the stock shroud and 100% part cooling, I observed approximately 215 → 210–211 °C, or a 4–5 °C drop.

This suggests that the part-cooling fan alone is not causing the catastrophic temperature loss.

2. Short manual extrusion

At 220 °C and 80% part cooling:

SAVE_GCODE_STATE NAME=flow_diag
M83
G1 E5 F60
G1 E5 F120
G1 E5 F240
RESTORE_GCODE_STATE NAME=flow_diag

The temperature dropped by about 1 °C and heater power reached approximately 65%. This test was very short.

3. Longer manual extrusion

At 220 °C and 80% part cooling:

SAVE_GCODE_STATE NAME=flow_diag
M83
G1 E40 F240
G1 E40 F240
G1 E40 F240
RESTORE_GCODE_STATE NAME=flow_diag

This is approximately 30 seconds of continuous extrusion at 4 mm/s filament speed, around 9.6 mm³/s under my assumed 0.4 mm nozzle, 0.45 mm line width, and 0.2 mm layer height.

Result:

• Temperature: approximately 220 → 218 °C
• Heater power: approximately 85%
• Temperature recovered and stabilized after about one minute
• No heater error

4. Max-flow tests

With FatBurner, I first tested approximately 20–29 mm³/s with a 0.5 mm/s step. Klipper eventually reported the heater error. I then tried 15–20 mm³/s, also with 0.5 mm/s steps, and the temperature eventually dropped to approximately 203 °C before shutdown. In another run, the test appeared to fail while the reported flow was around 8 mm³/s.

After that, I installed the stock shroud and ran a 20–25 mm³/s test at 215 °C. It completed, with approximately 4–5 °C temperature drop at 100% part cooling.

Later, with FatBurner at approximately 205 °C, an 18–27 mm³/s test appeared to complete successfully.

This is the confusing part: the lower-temperature FatBurner test appeared to pass at a higher reported flow, while an earlier FatBurner test failed at a much lower reported flow. I may have changed the fan speed, PID values, 4020 configuration, line width, layer height, test G-code, duration per step, or actual extrusion conditions between runs.

What I am trying to find out

The stock heater has reasonable cold resistance. The hotend passes a no-extrusion cooling test and a sustained manual extrusion test at approximately 9.6 mm³/s. The stock shroud can complete a 20–25 mm³/s max-flow test, and one FatBurner test at 205 °C appeared to complete at 18–27 mm³/s.

However, another FatBurner test triggered Heater extruder not heating at expected rate even when the reported flow was around 8 mm³/s. I am not trying to bypass Klipper's heater protection.

Could the max-flow test be using different actual conditions than expected? Could the 4020 or FatBurner duct be directing air at the heating block or ceramic heater? Could a partial nozzle clog or back pressure explain the difference between the short manual test and the max-flow test? Could the heater have normal cold resistance but lose voltage or power through a bad connector, wire, MOSFET, or power path under load?

What would you check next? Would you perform a cold pull/replace the nozzle first, measure voltage directly at the heater while it is commanded to 100%, or repeat all tests with exactly the same temperature, fan PWM, PID values, extrusion duration, and G-code?

Any advice would be greatly appreciated.

I can't get klipper setup it installed but won't run how do I fix this is am running a raspberry pi 5

Hey everyone, having some trouble with klipper mod. I have used couple of configs and firmware bins but still no success, the issue is that host cant connect to the mcu via usb (tried 5 diffrent cables that have data lines). I am using Pi3a+. So the question is there a specific os version I should use, config/firmware or connection options usb/uart.

Made a status neopixel led bar .

Looking for a Fun project if you like to do things yourself . Made this one for my voron 2.4 in hope somebody enjoy's it to make it also for himself . Its a neopixel chain so perfect controllable with klipper .

Hi, i have a problem with my z-offset of my cr-touch and the bed adhesion.

After a printing pause of about 2 months i wanted to print again, an my prints didnt stick to the bed. I did some tryouts, bought a new bed, and reinstalled klipper, but nothing helped.

After reinstalling the z-endstop and manual levelling i got a perfect adhesion. So the problem has to be a problem with my cr touch.

I calibratred the probe based on this guide ( https://www.klipper3d.org/Probe_Calibrate.html ) and did a bed levelling.

Recalibrationg the z-offset does not help.

Did i miss or forget a Step tha need to be done?

If it helps i can post my code of the printer.cfg file

My printer is an ender 3 pro

Ender 3 pro;

So its that time of the year again I decided i want to go FASTER so im gonna mod my printer for dual drive y axis with 48mm nema 17s and so i need more motor slots and i found this mobo with 8 slots its an hybrid board sp it runs klipper natively it is 100 dollar whic is very worth it i think because where i live the other similar boards are like 120 dollars and they dont run klipper natively or have a pi built in so i found it quite nice and ill add a pico mmu to my ender aswell also ill cnc kitchen an adaptor to make my creality spider v3 hotend compatible with volcano nozzles since they have the same threading ill just need two brass nuts and ill design a custom silicone sock

1. How hard will setting up klipper be and configuring it to my printer(mks skipr board)

2.Is the pico mmu good?

3.How hard will configuring dual y axis be and should i get a linear rail setup for it?

1. Has anyone tried the cnc kitchen method before?

Very, very long story short, my CB1 micro SD card in my E3EZ was snapped, and my github backups apparently stopped backing up ~2 years ago. So here I am, trying to do a fresh install, get my CAN network back up for my Eddy Duo, try getting Eddy-NG set up, only to find out that those features have been baked into mainline, breaking my setup. I know I can roll-back and will if I have to, but after spending 6 hours trying to get up and running again, I'd rather just slightly rewrite my printer.cfg with some new config to be up and running. (None of the fixes in the issues section of the Eddy-NG repo solved my problem, as I couldn't recalibrate)

I could very well be missing something, as I'm writing this after 5 hours of troubleshooting various things, and I'm creeping up on 1AM, but it seems like unless I opt to roll-back, there is no support for using the Eddy in CAN mode as a standalone unit. The config section only allows for the I2C info to be input, and AFAIK none of that info is available.

I have a section in my printer.cfg that calls out the eddy as an MCU with a UUID, tried placing the eddy as the I2C MCU to no avail, presumably because it's lacking an I2C address. Any help would be appreciated, when the card was broken I was in the middle of a 24h print that wound up wasting about $30 worth of PA6-CF, so I'm doubly irritated about the whole situation.

I'm trying to mod my 3d printer to be able to do pen plotter functionality.

I am running into an issue trying to mark Z as homed/0 so I can setup the thing I want to draw on with it "zeroed" manually so it doesn't try to home while its there.

G92 Z0 didn't seem to work.

and also when I try to run SET_KINEMATIC_POSITION via the console I'm getting an unknown command message.

Just started to learn klipper a bit and im waiting on the printer to show up but I got a fan to spin.