Aquilawire VS.807

Converting a Voxelab Aquila X2 into a CoreXZ “Switchwire-style” printer

2023-9-10 • Ryan H

TL;DR: I repurposed an Aquila X2 into a CoreXZ machine inspired by the VORON Switchwire to eliminate Z-wobble, add linear rails, and enable a modern toolhead/probing stack. Result: faster, cleaner motion and predictable first layers, with parts availability and serviceability in mind.

Problem

The stock bed-slinger was fine for learning, but inconsistent Z mechanics and ringing limited surface quality and throughput. I wanted rigidity, repeatability, and serviceability without binning the whole platform.

Constraints & goals

• Reuse the frame where sensible: keep overall footprint and salvage value.
• Mechanical rigidity: linear rails, CoreXZ kinematics, tidier belt paths.
• Modern toolhead: Stealthburner + CW2 for clean wiring and cooling.
• Probing & auto Z: hardware-repeatable Z offset and reliable mesh.
• Parts availability: off-the-shelf fasteners/electronics here in AU.
• Serviceability: sane electronics layout, labelled harnesses, laced cabling.

Solution overview

I followed the excellent Ender_SW approach for CoreXZ on an Ender/Aquila frame, then adapted the toolhead and probing to my needs. Linear rails handle X/Z, CoreXZ removes the typical lead-screw Z-wobble failure mode, and a Stealthburner + CW2 gives consistent part cooling and plug-in hotend options. For probing, I landed on KlickyPCB after testing alternatives, paired with Klipper macros for automatic nozzle offset.

Build notes

• Frame reuse: the Aquila/Ender extrusion set is a reasonable donor once squared; CoreXZ brackets take care of the heavy lifting.
• Rails & alignment: tram the rails to the gantry carefully; backlash here shows up as ripples.
• Probe path: KlickyPCB gave better repeatability for me than early KlickyNG prints; servo mount keeps the mechanism simple.
• Auto-Z offset: Klipper macros measure nozzle-to-probe offset per swap—no feeler gauge required.
• Electronics bay: give cables strain relief and route stepper pairs away from signal lines; label everything.

Printer Performance

• External walls: 300 mm/s @ 10000 mm/s²
• Travel: 400 mm/s @ 15000 mm/s²
• Resonance (input shaper): X ≈ 77.6 Hz, Y ≈ 29.2 Hz
• Materials validated: PLA, ABS/ASA, PETG, TPU (profiles tuned)

Images

Specifications

• Stealthburner w/ CW2
• Dragon UHF w/ CHT Volcano
• BTT SKR Mini E3 V3.0
• Raspberry Pi 4 (8 GB)
• CYD screen (custom case)
• KlickyPCB probe (servo mount)

Why not just buy a printer?

If your priority is time-to-part and vendor support, buying a ready CoreXY can make sense. This project was about learning, reuse, and tuning a platform—and proving what can be done with a modest donor and careful design and implementation.

Apply this approach

The same method—requirements first, constrained design, repeatable measurement—is what we bring to client work. If you need help selecting a machine, upgrading motion/tooling, or profiling materials, see our 3D Printing and Design & Drafting services, or get in touch.

FAQ

Build or buy—what makes sense?
If you value learning and customisation and can invest time, build. If you need predictable output now with minimal tuning, buy.

Why base it on VORON Switchwire ideas?
It’s a well-documented, rigid CoreXZ pattern with community-proven parts and good serviceability.

Why convert an Aquila instead of starting from scratch?
Lower cost, reuse of extrusions/hardware, and a compact footprint already on hand.

What does CoreXZ improve over a bed-slinger?
Better Z straightness (no lead-screw wobble), lower moving mass in the print plane, and cleaner ringing characteristics when tuned.

What did the upgrade change in real numbers?
Surface finish improved (ringing reduced), first-layer yield increased, and stable speeds rose to 300/400 mm/s with input shaping.