First-Layer Squish, Line Width, and Z-Offset Calibration

First-layer adhesion and bottom-surface dimensional accuracy depend on matching the physical nozzle-to-bed clearance (h_actual) to the initial layer height (h_1) commanded by your slicer. Because the slicer extrudes a fixed volume of plastic based on h_1 and the first-layer line width (w_1), any Z-offset error (Δz) directly alters the actual cross-sectional area available beneath the nozzle (w_effective = w_1 × h_1 / (h_1 + Δz)). Calibrating Z-offset after full thermal expansion at printing temperature—combined with a first-layer line width of 120% to 140% of nozzle diameter—maximizes bed contact pressure while preventing elephant’s foot and plowed surface ridges.

Nozzle-to-Bed Clearance and Volume Conservation Math

When the slicer generates G-code for the first layer (for example, G1 Z0.20), it assumes that Z = 0.00 mm represents the exact plane where the nozzle tip touches the build plate. If your probe or endstop Z-offset has an uncompensated error Δz (where negative Δz means the nozzle sits closer to the bed than commanded, and positive Δz means the nozzle sits higher above the bed), the true physical gap is:

h_actual = h_1 + Δz

Assuming your extruder and filament multiplier are tuned via Extruder E-Steps and Rotation Distance and Flow Rate and Extrusion Multiplier Calibration, the extruder delivers the exact volume of molten polymer calculated for h_1 and w_1. Because molten polymer is incompressible, squeezing that constant volume into a vertical clearance h_actual forces the effective extruded line width (w_effective) to scale inversely with height:

w_effective ≈ w_1 × (h_1 / h_actual) = w_1 × [h_1 / (h_1 + Δz)]

How Small Z-Offset Errors Amplify First-Layer Over-Extrusion

Consider a 0.40 mm (0.0157 in) nozzle printing a first layer at h_1 = 0.20 mm (0.0079 in) height and w_1 = 0.50 mm (0.0197 in) line width. A ±0.04 mm (±0.0016 in—less than half the thickness of standard copier paper) Z-offset error substantially changes effective line width and fill percentage:

Z-Offset Error (Δz)	True Nozzle Gap (h_actual)	Effective Line Width (w_effective)	Space-Filling Ratio (h_1 / h_actual)	Physical Behavior on Build Plate
+0.08 mm (Too high)	0.28 mm	0.357 mm	71.4% (-28.6% fill)	Round, unflattened cord; zero lateral contact; detaches on corners
+0.04 mm (Slightly high)	0.24 mm	0.417 mm	83.3% (-16.7% fill)	Parallel gaps between infill lines; visible valleys on bottom face
0.00 mm (Calibrated)	0.20 mm	0.500 mm	100.0% (Exact fill)	Uniform flat bottom; lines fuse cleanly without upward ridges
-0.04 mm (Slightly low)	0.16 mm	0.625 mm	125.0% (+25.0% fill)	Excess plastic plows up into sharp ridges between adjacent passes
-0.08 mm (Too low)	0.12 mm	0.833 mm	166.7% (+66.7% fill)	Translucent wave patterns, severe elephant’s foot, extruder clicking

This inverse relationship (h_1 / (h_1 + Δz)) explains why the Slic3r First Layer manual recommends a generous first-layer height (0.20 to 0.25 mm on a 0.40 mm nozzle, even when upper layers print at 0.10 to 0.12 mm). On a 0.25 mm first layer, a -0.03 mm bed flatness ripple creates a manageable 13.6% local volume excess (0.25 / 0.22 = 1.136). On a 0.10 mm first layer, that same -0.03 mm ripple creates a 42.9% volume excess (0.10 / 0.07 = 1.429), causing severe nozzle scraping.

Sign Conventions in Marlin (M851) vs. Klipper (z_offset)

Verify how your firmware defines the sign of its probe offset before making adjustments:

• Marlin (M851 Z): Defines the probe trigger point relative to the nozzle (Z_probe_trigger - Z_nozzle). Because an inductive or BLTouch probe triggers below the nozzle tip, M851 Z is negative (such as M851 Z-1.85). Making M851 Z more negative (-1.90) moves the nozzle closer to the bed (-0.05 mm).
• Klipper ([probe] z_offset): Defines the distance the nozzle must descend after the probe triggers to reach Z = 0. For a probe that triggers above the bed, Klipper’s z_offset in printer.cfg is positive (z_offset: 1.850). Increasing z_offset to 1.900 moves the nozzle closer to the bed, or you can run SET_GCODE_OFFSET Z_ADJUST=-0.05 MOVE=1 during a live print followed by Z_OFFSET_APPLY_PROBE.

Thermal Expansion of the Hotend and Build Plate

Calibrating Z-offset with paper at room temperature (20 °C / 68 °F) and immediately printing at high temperature leaves the nozzle closer to the bed than measured. Metallic components in the vertical loop expand according to linear thermal expansion:

ΔL = α × L₀ × ΔT

Where ΔL is length change (mm), α is the coefficient of linear thermal expansion (10⁻⁶ /K), L₀ is initial heated length (mm), and ΔT = T_print - T_ambient (K or °C).

Worked Calculation: Room Temperature vs. Printing Temperature

Consider the vertical expansion of a direct-drive hotend and cast aluminum bed when heating from 20 °C (68 °F) to ABS/ASA printing temperatures (250 °C / 482 °F nozzle, 100 °C / 212 °F bed; see PLA vs. PETG vs. ASA Mechanical and Thermal Properties):

1. Brass Nozzle + Heater Block (α_brass ≈ 19 × 10⁻⁶ /K, L₀ = 20.0 mm, ΔT = 230 K):

   ΔL_hotend = 19 × 10⁻⁶ × 20.0 mm × 230 K = 0.0874 mm (87.4 µm)

2. Stainless Steel Heatbreak Transition (α_ss ≈ 16 × 10⁻⁶ /K, L₀ = 12.0 mm, average ΔT = 110 K):

   ΔL_heatbreak = 16 × 10⁻⁶ × 12.0 mm × 110 K = 0.0211 mm (21.1 µm)

3. Aluminum Bed Plate + PEI Sheet (α_al ≈ 23 × 10⁻⁶ /K, effective thickness above mounts L₀ = 8.0 mm, ΔT = 80 K):

   ΔL_bed = 23 × 10⁻⁶ × 8.0 mm × 80 K = 0.0147 mm (14.7 µm)

Summing these expansions (0.0874 + 0.0211 + 0.0147 = 0.1232 mm) reduces nozzle-to-bed clearance by 0.12 mm (123 µm) between cold and hot states—more than half of a 0.20 mm first layer. As noted in the Klipper Probe Calibration guide, standard 80 gsm copy paper is roughly 0.10 mm thick, which coincidentally offsets hotend thermal growth when calibrated cold. However, for repeatable results across PLA and ASA bed temperatures, pre-heat the bed for 5 to 15 minutes (with the nozzle at a 150 °C non-oozing standby temperature) before homing and bed meshing, then fine-tune the final 0.02 to 0.04 mm on a live first-layer patch.

Why Wider First-Layer Lines (120%–140% of Nozzle Diameter) Improve Adhesion

Slicer profiles typically set first-layer line width (w_1) between 120% and 140% of nozzle diameter d_n—such as 0.48 mm to 0.56 mm on a 0.40 mm nozzle (compare width-to-height combinations in the Flow & Volumetric Speed Calculator and Nozzle Size vs. Layer Height Limits).

1. Exploiting the Nozzle Tip Annular Flat: A 0.40 mm nozzle orifice is surrounded by a flat brass ring (d_flat ≈ 0.80 to 1.00 mm). Commanding a 0.50 to 0.56 mm line width forces molten polymer laterally under this flat shoulder, which acts like a heated trowel to press plastic into the micro-texture of smooth or textured PEI.
2. Higher Wetting Pressure Without Reducing Z-Clearance: Widening w_1 increases contact area and melt-pool pressure while maintaining a safe 0.20 mm vertical clearance above bed high spots.

Diagnosing First-Layer Surface Defects and Elephant’s Foot

Printing a single-layer 60 × 60 mm (2.36 × 2.36 in) square at 0.20 mm height provides a direct visual and tactile diagnostic, as shown in the Prusa First Layer Calibration guide:

Visual & Tactile Symptom on Single-Layer Patch	Underlying Clearance Cause	Recommended Z-Offset Adjustment
Rounded strings; gaps between lines; sheet pulls apart easily when peeled	Nozzle too high (Δz > +0.03 mm); stadium edges fail to meet	Lower nozzle in -0.02 mm to -0.04 mm increments
Center lines touch, but tiny pinholes remain where infill meets perimeter	Nozzle slightly high (Δz ≈ +0.01 to +0.02 mm)	Lower nozzle by -0.01 mm to -0.02 mm
Smooth, flat top and bottom; fuses into a tear-resistant sheet measuring ~0.20 mm	Ideal clearance (Δz ≈ 0.00 mm)	Save offset (M500 or SAVE_CONFIG)
Rough, sandpaper-like ridges along line boundaries; fingernail catches across paths	Nozzle slightly too low (Δz ≈ -0.02 to -0.05 mm); nozzle flat plows excess plastic upward	Raise nozzle in +0.02 mm increments
Translucent wavy ripples; extruder clicks; wide flare on bottom perimeter	Nozzle severely too low (Δz < -0.06 mm); orifice obstructed by bed	Raise nozzle by +0.05 mm to +0.10 mm

Controlling Elephant’s Foot for Precision Assemblies

Even at an ideal Z-offset, first-layer squish pressure and bed heat cause the bottom 1 to 2 layers to flare outward by 0.10 to 0.20 mm per side (elephant’s foot), interfering with mating parts and vertical bores (see 3D Print Clearances and Fits and Why 3D Printed Holes Come Out Undersized). Rather than raising Z-offset and sacrificing adhesion, enable Elephant foot compensation (0.10 to 0.15 mm) in your slicer to inset the first layer’s XY contour, and verify mating tolerances in the Fit & Tolerance Calculator.