最佳 Voron Z 探针 — Klicky vs Euclid vs Tap vs Sensorless vs Inductive

探头 电子 对比

The Z probe is one of the most debated topics in the Voron community. It directly affects your first-layer consistency, bed mesh quality, and overall print reliability. A bad probe means failed prints, damaged PEI sheets, and hours of frustration. A good probe means set-it-and-forget-it bed leveling for months. 最后更新：2025 年 5 月。

本指南对比五种 Z 探测方法：Inductive 探针 (PL-08N, TL-Q5MC)、Klicky (对接机械开关)、Euclid (磁性对接探针)、Tap (直接喷嘴接触) 和 Sensorless Homing (堵转检测)。涵盖微米级重复精度、Z 偏移稳定性、温度敏感性、成本、安装难度、工具头兼容性及常见故障模式。

概览 — 探针对比表

Tap — 新黄金标准

Tap 是 Voron 社区设计的探针，它将整个 Stealthburner 工具头下部替换为垂直滑动托架。喷嘴直接接触热床，推动整个托架向上移动，直至机械限位开关（微动开关或光学传感器）触发。这意味着 Tap 测量的是真实的喷嘴到热床距离，无偏移、无温度漂移、无长期校准漂移。

优点： Unmatched repeatability — ±2 µm is typical, and many users report consistent probes within ±1 µm. No Z offset drift because there is no offset to calibrate — the nozzle is the probe. Not affected by temperature changes (the nozzle doesn't change length meaningfully with heat). Extremely reliable — the single moving part is the sliding carriage on linear rails. Can probe hot or cold beds (inductive probes fail on cold PEI). Print start macros are simpler because you don't need to dock and undock the probe.

缺点： Stealthburner-only — you cannot use Tap with Afterburner, Mini Stealthburner, or other toolheads. The sliding carriage adds approximately 45g to the toolhead mass, which reduces acceleration limits and requires retuning input shaper. The probe action uses the Z motor to push the nozzle into the bed — if your Z endstop or probe trigger fails, you can crash the nozzle into the bed. There is a minimum Z hop requirement (typically 5-10mm) to clear the probe trigger point.

最佳适用： Voron V2.4 and Trident builds that can handle the extra toolhead mass. Tap is the most accurate and maintenance-free probing method available. If you want the best first layers possible and don't mind tuning input shaper after installation, Tap is the clear winner.

Tap Klipper 配置

[probe]
pin: ^!PA7                  ; Check your board pin mapping
x_offset: 0.0               ; Tap has zero X/Y offset from nozzle
y_offset: 0.0
z_offset: 0.0               ; Calibrated via PROBE_CALIBRATE
speed: 5.0                  ; Probing speed in mm/s
lift_speed: 10.0            ; Lift speed after probe
samples: 3
samples_result: median
sample_retract_dist: 3.0
samples_tolerance: 0.005    ; 5µm tolerance between samples
samples_tolerance_retries: 3

[stepper_z]
endstop_pin: probe:z_virtual_endstop
position_endstop: 0.0
    

Klicky — 经济型机械精度

Klicky is a dockable mechanical probe that uses a microswitch mounted on a 3D-printed arm. The probe docks on a printed holster mounted to the frame or gantry. When homing or probing, the toolhead moves to the dock, picks up the probe via a magnet on the probe body, performs the probe, and returns the probe to the dock. It was the first community-designed Voron probe and remains extremely popular.

优点： Extremely low cost — you need a microswitch ($1-2), a magnet ($1-2), printed parts, and a small PCB or wiring. Excellent repeatability for a mechanical switch — ±5 µm is realistic. Compatible with every Voron toolhead (Stealthburner, Afterburner, Mini Stealthburner, V0.2, Switchwire) because the probe mounts to the toolhead body, not inside it. Temperature stable — a microswitch triggers at the same force regardless of chamber temperature. Well-documented with years of community refinements.

缺点： More complex setup than Tap — you need to configure docking macros, probe pick-up routines, and verify that the probe docks reliably every time. The printed arm can wear out over hundreds of dock cycles, requiring re-printing. Mis-docking is the most common failure — if the probe isn't seated fully on the magnet, the probe position is off by 1-2mm. Filament strings can interfere with docking if the nozzle isn't wiped clean before probing. The dock takes up space on the frame and can interfere with cable chains on some builds.

最佳适用： V0.2 and Switchwire builds where Tap doesn't fit, and Trident builds where you want better accuracy than inductive probes but don't want to add toolhead mass. Also excellent for budget builds — you can build a Klicky for under $10 in parts.

Klicky Klipper 配置

[probe]
pin: ^!PC15                 ; Check your board pin mapping
x_offset: 0.0               ; Klicky has zero X/Y offset from nozzle
y_offset: 0.0
z_offset: 0.0               ; Calibrated via PROBE_CALIBRATE
speed: 20.0                 ; Probing speed
lift_speed: 15.0
samples: 3
samples_result: median
sample_retract_dist: 5.0
samples_tolerance: 0.0075   ; 7.5µm tolerance
samples_tolerance_retries: 3

[gcode_macro _PROBE_PICKUP]
gcode:
    ; Move to dock position - adjust coordinates for your build
    G90
    G1 X150 Y250 F12000
    G4 P500
    ; Pick up probe
    PROBE_DOCKED=1

[gcode_macro _PROBE_DOCK]
gcode:
    ; Return probe to dock
    G90
    G1 X150 Y250 F12000
    G4 P500
    PROBE_DOCKED=0
    

Euclid — 磁性精准探测

Euclid is an evolution of the Klicky concept that replaces the microswitch arm with a magnetic docking body. The probe consists of a magnetically-attached body containing a microswitch, which docks onto a magnetic receiver on the toolhead. The key difference from Klicky is that Euclid uses a dedicated PCB and a more refined magnetic latching mechanism for more consistent docking.

优点： More consistent docking than Klicky — the dual-magnet latch system is less sensitive to positioning errors. The dedicated PCB with status LED helps with debugging. Higher quality microswitch than typical Klicky builds (Omron D2F-5L instead of generic limit switches). Better Z offset stability than inductive probes. The probe body is more rigid than the printed Klicky arm, reducing flex-induced variation. Good community documentation and config templates.

缺点： More expensive than Klicky — the PCB and magnets add up to $12-20 per probe set. Still requires docking macros and careful tuning for reliable pick-up/drop-off. The probe body adds about 8g to the toolhead when docked, which is less than Tap's 45g but still measurable on input shaper tuning. The magnetic connection can lose strength over time if the magnets are exposed to high chamber temperatures (above 70°C). Still uses a microswitch, which has a finite lifespan (~1 million actuations).

最佳适用： Trident and V0.2 builds where you want better reliability than Klicky without the toolhead mass of Tap. Euclid is also a good upgrade path for builders who started with Klicky and want more consistent docking behavior.

Inductive 探针 — PL-08N 和 TL-Q5MC

电感式探针是 Voron 最初的探测方法。这些是简单的传感器，用于检测金属（您的 PEI 弹簧钢片或铝制热床）的存在。最常见的型号是 Omron PL-08N（NPN 常开）和 TL-Q5MC（PNP 常闭）。它们通过打印支架安装在工具头上，当热床表面进入传感器面约 2-4mm 范围内时触发。

优点： Simplicity — no moving parts, no docking, no macros. Wire the sensor to an endstop pin, configure a probe section in Klipper, and probe. Cheap — $3-10 for genuine Omron or $1-2 for generic clones. Works with any toolhead that has a probe mount. No toolhead mass added beyond the sensor itself (~15g). Easy to troubleshoot — if it triggers on a piece of metal, it works.

缺点： Temperature sensitivity is the killer. Inductive probes drift significantly with temperature — as the chamber heats up from 25°C to 60°C, the Z offset can shift by 30-100 µm due to thermal expansion of the probe mount and sensor electronics. This means your first layer calibration at room temperature is wrong when the chamber is hot. The probe must be mounted at a precise distance from the nozzle (typically 2-4mm offset in Z and 20-30mm offset in X/Y), creating a lever arm that amplifies any tilt in the toolhead. The probe only works on metal build plates — a G10/FR4 or glass bed won't trigger it. The 4mm sensing distance means the nozzle can crash into the bed if the probe fails.

最佳适用： Legacy Voron builds that already have inductive probes installed and working. If you're building a new Voron in 2025, we strongly recommend Tap, Klicky, or Euclid instead. The temperature drift issue alone is enough reason to skip inductive probes.

Inductive 探针 Klipper 配置 (PL-08N)

[probe]
pin: ^!PA15                 ; PL-08N NPN normally open
x_offset: 20.0              ; X offset from nozzle to probe center
y_offset: 0.0               ; Y is typically on center
z_offset: 0.0               ; Calibrated via PROBE_CALIBRATE
speed: 10.0
lift_speed: 15.0
samples: 3
samples_result: median
sample_retract_dist: 5.0
samples_tolerance: 0.010    ; 10µm tolerance (wider due to drift)
samples_tolerance_retries: 3
    

Sensorless Homing — 零成本方案

Sensorless homing (also called stall detection or TMC homing) uses the TMC2209/TMC2240/TMC5160 stepper driver's StallGuard feature to detect when the axis physically hits an obstruction (the bed or a hard stop). No probe hardware is needed — the driver senses the motor's load and triggers the endstop when the load spikes from the collision. This is commonly used for X and Y homing on Vorons, and some builders use it for Z probing as well.

优点： Zero hardware cost — it's a feature of the TMC2209/2240/5160 drivers you already have. Zero toolhead mass. No probe mounting, wiring, or maintenance. Works on any build plate material (metal, PEI, G10, glass). No moving parts to wear out. Useful as a backup homing method even if you have a dedicated probe.

缺点： Repeatability is poor compared to dedicated probes — ±15-30 µm is typical, and it varies with temperature, bed surface condition, and axis speed. Requires careful tuning of the stall sensitivity threshold (driver_SGTHRS) — too sensitive triggers early, too insensitive crashes the nozzle into the bed with more force. Not suitable for automatic bed mesh probing — the variance is too high for a useful mesh. The collision force can damage PEI sheets or the nozzle over many cycles. The Z motor must be strong enough to lift the gantry against the bed — a TMC2209 at 1.2A may not reliably detect a stall against a PEI sheet.

最佳适用： Backup homing method in combination with a primary probe (Tap, Klicky, or Euclid). Many Voron users run sensorless X and Y homing (to eliminate endstop switches) with a Tap or Klicky for Z probing. Using sensorless as the sole Z probing method is not recommended for a daily driver printer.

Sensorless Homing Klipper 配置 (Z 轴)

[stepper_z]
endstop_pin: tmc2209_stepper_z:virtual_endstop   ; Uses driver stall detection
position_endstop: 0.0
homing_speed: 5.0                                 ; Slow speed for reliable detection
homing_retract_dist: 3.0

[tmc2209 stepper_z]
diag_pin: ^!PD2                                   ; Check your board pin mapping
driver_SGTHRS: 70                                 ; Stall sensitivity - tune this!
    

常见探针问题排查

Tap — 误触发

Tap false triggers are usually caused by carriage binding or excessive grease. The sliding carriage needs to move freely — if it binds (too-tight eccentric nuts, misaligned linear rails, or hardened grease), the carriage may not return to its home position after probing. Fix: clean and re-lubricate the linear rails, loosen the eccentric nut slightly, and verify that the carriage drops under its own weight when the toolhead is removed from the printer. False triggers can also occur if the Z homing speed is too high — keep it at 5mm/s or below.

Klicky/Euclid — 对接失败

对接失败是 Klicky 最常见的故障。探针通过磁力吸附在工具头上，但未正确就位，导致探针倾斜。修复方法：使用千分表或直角尺确保工具头和停靠位精确对齐。增加磁铁强度（使用 N52 而非 N48 磁铁）。在停靠位的导向面上增加倒角，使探针能自动居中。清除停靠位和探针磁面上的任何耗材碎屑。如果问题持续，请检查您的龙门架方正度——倾斜的龙门架会使工具头相对于固定停靠位错位。

Inductive — 温度漂移

如果您的第一层在 25°C 时完美，但在腔室温度达到 60°C 时变得过高，则您的电感式探针正在漂移。修复方法是在工作温度下校准 Z 偏移：将腔室加热到您的典型打印温度（60°C），让所有部件热平衡 15 分钟，然后运行 PROBE_CALIBRATE。您也可以使用 Z 限位开关进行归零，仅将探针用于热床网格探测——这样将归零 Z 偏移（必须稳定）与网格（可容忍轻微漂移）分开。更好的方法是切换到 Tap。

Sensorless — 检测不一致

如果无传感器归零在不同距离触发，请调校 driver_SGTHRS 值。较低的值使检测更灵敏（更早触发），较高的值使检测更不灵敏（更晚触发，力量更大）。从 70 开始，以 5 为增量调整。最佳值取决于您的 Z 轴摩擦力、电机电流和龙门架重量。每次调整后运行 PROBE_ACCURACY 以测量重复性。如果无法将变化控制在 ±25µm 以下，请切换到专用 Z 探针。

按 Voron 型号推荐

Voron V2.4： Tap. The V2.4's gantry is heavy enough that the extra 45g of Tap makes a negligible difference to acceleration limits. The unmatched repeatability (±2µm) ensures perfect first layers across the full 350mm print area. Pair with sensorless X and Y homing for a fully endstop-free build.

Voron Trident： Tap or Euclid. Tap if you want the best possible probing accuracy. Euclid if you want to keep toolhead mass low (the Trident's moving bed means the toolhead mass directly affects print quality at high speeds). Both are excellent choices.

Voron V0.2： Klicky or Euclid. Tap doesn't fit the V0.2's Mini Stealthburner toolhead. Klicky is the budget option; Euclid is the refined option. Both give good accuracy without adding significant mass to the tiny V0.2 toolhead.

Voron Switchwire： Klicky. The Switchwire's cantilevered gantry is sensitive to toolhead mass, so Tap is not ideal. Klicky's low cost and good accuracy make it the natural choice. Sensorless Z can work on the Switchwire due to the lightweight bed, but a mechanical probe is more reliable.

预算推荐： Klicky. For under $10, you get probing accuracy within ±5µm, temperature stability, and compatibility with every Voron model. It requires more setup than Tap, but the cost savings are significant.