ADR-014: Z-Probe Selection (SuperPINDA vs BLTouch)
Status
Accepted
Context
The Amalgam requires a Z-probe for homing and bed-leveling. The choice depends heavily on the bed surface material:
Two Probe Technologies: 1. BLTouch: Physical Hall Effect probe with moving pin that touches bed 2. SuperPINDA: Inductive proximity sensor that detects metal without contact
Bed Surface Constraints: - Inductive Probes: Only work on conductive metal (magnetic spring steel sheets like MK52) - Physical Probes: Work on any surface (glass, spring steel, wood, carbon fiber, plain aluminum)
Engineering Philosophy: - “Tractor” Goal: Set-and-forget reliability with minimal maintenance - “Scavenger” Goal: Universal compatibility with repurposed beds
This creates a decision matrix based on both bed surface and reliability requirements.
Decision
We adopt a bed-surface-dependent probe strategy with SuperPINDA as Reference Spec and BLTouch as Scavenger Fallback.
Probe Options
Option A: SuperPINDA (Reference Spec) - Technology: Inductive proximity sensor (solid-state) - Bed Requirement: Magnetic spring steel sheet (MK52 or compatible) - Mounting: Fixed position on X-carriage or gantry - Wiring: 3 wires (VCC, Ground, Signal) - Reliability: Maximum - no moving parts, resin-encased, nearly indestructible - Cost: ~$30-40 AUD - Best For: Tier 3 Reference Spec builds with MK52 bed
Option B: BLTouch V3.1 (Scavenger Fallback) - Technology: Physical Hall Effect probe with deployable pin - Bed Requirement: Any surface (glass, carbon fiber, aluminum, spring steel) - Mounting: Fixed position on X-carriage or gantry - Wiring: 5 wires (VCC, Ground, Signal, Control Servo) - Reliability: Moderate - moving pin can snap or bend if it hits print/clip - Cost: ~$50-60 AUD (authentic), ~$15-25 AUD (3DTouch clone) - Best For: Tier 1-2 builds with scavenged beds (glass, etc.)
Technical Comparison
| Feature | SuperPINDA (Inductive) | BLTouch (Physical) |
|---|---|---|
| Bed Surface | Metal only (conductive) | Universal (any material) |
| Accuracy | Extremely high, repeatable | Extremely high, but can drift if pin bent |
| Robustness | Maximum (solid-state, no moving parts) | Moderate (pin can snap/bend) |
| Complexity | Simple (3 wires, always-on) | Moderate (5 wires, deploy/retract logic) |
| Wear Items | None | Pin, springs, servo |
| Thermal Drift | Minimal (modern SuperPINDA) | None (mechanical contact) |
| Calibration | Z-offset once | Z-offset once, pin deployment verification |
Why SuperPINDA for Reference Spec?
1. Solid-State Reliability - Zero moving parts to fail - Immune to “pin-snag” failures during high-speed travel - Resin-encased design is nearly indestructible - No deployment mechanism to jam or misfire
2. Thermal Stability - Older inductive probes drifted with bed temperature - PINDA V2 added thermistor to compensate (0.02mm adjustments) - SuperPINDA uses higher-quality components that don’t drift - Trigger height identical at 20°C or 60°C chamber temperature
3. “Set-and-Forget” Philosophy - Once calibrated, requires no ongoing maintenance - No pin to straighten or replace - No servo mechanism to lubricate or adjust - Perfect for 1000+ day maintenance intervals
4. Simplicity - 3-wire connection (vs 5 wires for BLTouch) - No deployment/retract logic in Klipper - Always-on proximity sensing - Lower failure surface area
Why BLTouch for Scavenger Builds?
1. Universal Compatibility - Works on any bed surface (glass, carbon fiber, plain aluminum) - Only option for salvaged Ender 3, i3 Mega, Anet A8 beds - Allows flexibility in bed choice during Tier 1 builds - Doesn’t force MK52 bed purchase
2. Well-Understood Technology - Years of community documentation and troubleshooting - Wide availability of clones and spare parts - Proven track record in millions of printers
3. Bed-Surface Flexibility - User can experiment with different bed surfaces - No commitment to magnetic spring steel upfront - Universal “key” for uncertain scavenger inventory
Build Wizard Logic
Configuration Wizard → Z-Probe Selection
├─ Question: "What is your bed surface material?"
│ ├─ Glass / Carbon Fiber / Plain Aluminum
│ │ └─ Force: BLTouch (inductive cannot see)
│ ├─ Magnetic Spring Steel (MK52 or compatible)
│ │ └─ Recommend: SuperPINDA ★ Reference Spec
│ │ └─ Option: BLTouch (if preferred)
│ └─ Unknown / Not sure
│ └─ Default: BLTouch (universal fallback)
Bed-Probe Compatibility Matrix
| Bed Surface | SuperPINDA | BLTouch | Recommended |
|---|---|---|---|
| Magnetic Spring Steel (MK52) | ✅ Works | ✅ Works | SuperPINDA ★ |
| Plain Aluminum | ✅ Works | ✅ Works | BLTouch (easier) |
| Glass | ❌ Won’t detect | ✅ Works | BLTouch only |
| Carbon Fiber | ❌ Won’t detect | ✅ Works | BLTouch only |
| PEI on Glass | ❌ Won’t detect | ✅ Works | BLTouch only |
| BuildTak on Aluminum | ❌ Won’t detect | ✅ Works | BLTouch only |
Consequences
Benefits
- Tiered Flexibility: Reference path (SuperPINDA) and Scavenger path (BLTouch) both supported
- Bed-Surface Logic: Clear decision matrix based on bed material
- Reliability Optimization: Reference spec maximizes solid-state reliability
- Universal Fallback: BLTouch works with any scavenged bed
- Cost Awareness: Clones available for low-budget builds
Trade-offs
- SuperPINDA: Locked to metal beds, no flexibility in surface choice
- BLTouch: Higher failure rate (moving parts), pin maintenance required
- Cross-Compatibility: Cannot easily switch probes without changing bed
- Clone Quality: BLTouch clones vary in reliability (3DTouch vs authentic)
What This Enables
- Tier 3: SuperPINDA + MK52 bed = maximum reliability (reference spec)
- Tier 1-2: BLTouch + scavenged glass bed = universal compatibility
- All Tiers: Z-tilt calibration and mesh bed leveling (both probes support)
What This Replaces
- Single-probe assumption (no longer viable for mixed bed inventory)
- Glass-bed + inductive probe combinations (won’t work)
- Universal probe recommendation (bed-surface dependent)
BOM Implications (Generic)
Tier 3: Reference Spec (SuperPINDA + MK52)
- Parts needed:
- 1x SuperPINDA probe (Prusa or high-quality clone)
- 1x MK52 magnetic spring steel sheet (250x250mm)
- 1x Magnetic build surface (under spring steel)
- 1x Probe mounting bracket
- 3x Female-to-female dupont cables (VCC, GND, Signal)
- Cost implication: Low-Medium (~$50-70 AUD for probe + bed)
- Bed compatibility: Metal only (spring steel required)
- Reliability: Maximum (solid-state, no moving parts)
- Maintenance: None (set-and-forget)
Tier 1-2: Scavenger Fallback (BLTouch + Glass)
- Parts needed:
- 1x BLTouch V3.1 (authentic) or 3DTouch clone
- 1x Scavenged bed (glass, etc.)
- 1x Probe mounting bracket
- 5x Female-to-female dupont cables (VCC, GND, Signal, Control, Servo)
- Cost implication: Low (~$15-60 AUD depending on authenticity)
- Bed compatibility: Universal (any surface)
- Reliability: Moderate (moving parts, potential pin failure)
- Maintenance: Periodic pin inspection, occasional replacement
Hybrid Option: BLTouch on Metal Bed
- Use Case: User has metal bed but prefers BLTouch familiarity
- Cost implication: Same as BLTouch + metal bed (~$50-70 AUD)
- Trade-off: Loses SuperPINDA reliability advantages
- Recommended: Only if user has strong BLTouch experience or spare parts
Universal Requirements (All Options)
- Z-Max Switch: Required regardless of probe choice (see ADR-013)
- Mounting Bracket: 3D-printed bracket to secure probe to X-carriage
- Klipper Configuration:
[probe]or[bltouch]section based on choice - Z-Offset Calibration: Required for both probes (probe to nozzle distance)
Implementation Notes
SuperPINDA Klipper Configuration
[probe]
pin: ^PC2 # Example pin, adjust for board
x_offset: -10
y_offset: 20
z_offset: 1.6 # Calibrated via PROBE_CALIBRATE
samples: 3
sample_retract_dist: 2.0
speed: 5.0
[bed_mesh]
mesh_min: 30,30
mesh_max: 220,220
probe_count: 3,3BLTouch Klipper Configuration
[bltouch]
sensor_pin: ^PC2
control_pin: PA2
x_offset: -10
y_offset: 20
z_offset: 1.6
stow_on_each_sample: False
probe_with_touch_mode: True
[safe_z_home]
home_xy_position: 150,150
speed: 50
z_hop: 10Probe Calibration Procedure
# Home all axes
G28
# Calibrate Z-offset
PROBE_CALIBRATE
# Test repeatability
PROBE_ACCURACY
# (BLTouch only) Test pin deployment
BLTOUCH_DEBUG COMMAND=pin_down
BLTOUCH_DEBUG COMMAND=pin_upMounting Considerations
SuperPINDA: - Mount ~5mm above bed surface when deployed - Ensure bed is flat (spring steel on magnetic surface) - No need for retraction mechanism - Secure firmly (no vibration or movement)
BLTouch: - Mount so pin deploys ~2-3mm above bed - Test pin deployment before first print - Verify pin retraction doesn’t collide with bed clips - Check for pin bending after print failures
Troubleshooting
SuperPINDA: - Issue: Probe doesn’t trigger - Check bed is metal (spring steel) - Verify probe is within 4mm of bed surface - Test with multimeter (check signal pin toggles) - Issue: Inconsistent readings - Clean bed surface (metal debris can affect inductive) - Check wiring for loose connections - Verify probe mounting is secure
BLTouch: - Issue: Pin doesn’t deploy - Check servo control wiring - Verify [bltouch] section in config - Test pin manually with BLTOUCH_DEBUG commands - Issue: Pin stuck in deployed position - Emergency retract: BLTOUCH_DEBUG COMMAND=reset - Check for debris or bent pin - May need to disassemble and clean
Safety Considerations
- BLTouch Pin Breakage: Can occur if nozzle crashes into bed or print curls up
- SuperPINDA Distance: Too far from bed = won’t trigger; too close = premature trigger
- Wiring: Use shielded cables to prevent EMI interference
- Hotend Temperature: BLTouch pin can deform near hot nozzle (ensure proper mounting)
References
- docs/reference/ai-conversations/pinda-v-bltouch.md: Complete probe comparison
- docs/adr/013-drivers-endstops.md: Driver and endstop strategy
- docs/adr/005-triple-z.md: Triple-Z kinematic leveling
- Klipper Probe Guide: Probe Bed Calibrate
- Klipper BLTouch Guide: BLTouch
- Prusa SuperPINDA Documentation: Prusa Knowledge Base
Evolution Notes
This ADR establishes SuperPINDA as the Reference Spec probe for maximum reliability. Future probe technologies will be evaluated against: solid-state reliability, bed-surface compatibility, thermal stability, and cost. BLTouch remains the universal fallback for scavenger builds with non-metal beds.
Emerging Technologies: - LVDT sensors (high-precision, expensive) - Capacitive sensors (work on some non-metal surfaces, experimental) - Optical sensors (complex, not widely adopted)
Current consensus favors inductive (SuperPINDA) for metal beds and physical (BLTouch) for universal compatibility.