re:3D GigabotX 2 XLT 900 Power Assessment at JIFX 26-3
EQ-TR-2026-001 · May 2026 · Version 1.0
Continuous Waveform Measurement during Printing Operations
Kevin Davies, PhD · EQ Systems Inc.
Bottom Line
Every aspect of the printer’s electrical operation, observed over a 24.8-hour window spanning 12 printed parts, was found to be consistent with standard, best-practice engineering: controlled, zero-cross-aligned switching of the main heater; tight regulator-driven duty cycling; smooth motion-side current through the internal DC bus; clean supply within ANSI C84.1 Range A for more than 99% of the window; and per-part energy in the expected band for the machine class.
No observable operational concerns: peak current ≈ 10.8 A on a 20 A circuit, with substantial headroom.
The McMillan hangar grid feed itself measured cleanly: median 121 V, frequency stable at 60.005 ± 0.05 Hz, voltage THD under 2% (well under the IEEE 519 5% limit), and a ~0.37 Ω source impedance at the printer’s outlet, a stiff, quiet supply for a field facility.
The work validates comprehensive equipment assessment from a single point of measurement, from load disaggregation to supply characterization across time scales from 32 ksps waveforms through daily energy budgets, using continuous point-on-wave (CPOW) metrology bench-calibrated against a 6½-digit reference DMM to ~1% accuracy on current and under 0.5% on voltage at the loads observed.
Headline Findings
- Five distinguishable operating modes inferred from the AC power signature: a cross-product of heater state (off-duty vs full-on burst) and motion state (steppers and extruder de-energized vs energized), plus an idle baseline. Levels: ~52 W, ~210 W, ~300 W, ~1.04 kW and ~1.19 kW.
- Energy consumed: 4.08 kWh across 12 parts over the 24.8 h window: one long part in Session 1 (1.09 kWh, 139 min) and 11 shorter parts in Session 2 (mean 0.27 kWh, range 0.087–0.74 kWh; 2.99 kWh total).
- Peak draw ≈ 1.30 kW at the heated bed during a printing burst; running RMS stays well under 11 A over longer windows.
- Supply within ANSI C84.1 Range A for more than 99% of the window. The deepest heater-burst transient briefly dipped to 113.0 V (0.942 pu), well above the 0.9 pu (108 V) IEEE 1159 sag threshold. Frequency was exceptionally stable at 60.005 ± 0.05 Hz.
- Source impedance ≈ 0.37 Ω at the printer’s outlet (the V-vs-I slope at the plug), giving a steady voltage droop of ~0.37 V per amp of additional draw.
- The printer’s AC-DC supply isolates motor switching from the AC inlet. The NEMA 23 steppers are PWM-chopped on the printer’s internal 24 VDC bus, and that switching is filtered out before reaching the inlet. Motion activity therefore shows only as a smoothly varying base-load lift in heater-on-duty mode, with no stepper-PWM spectral content upstream.
Provenance
- Sensor: EQ Wave™ Waveform Measurement Unit (WMU) v1.1 hardware, 32 ksps CPOW plus 5 Hz power-monitor (PMon) aggregates, bench-calibrated against a 6½-digit reference DMM (current and voltage channels).
- Asset: re:3D GigabotX 2 XLT 900 large-format fused-granulate-fabrication (FGF) printer (U.S. Army property, manufactured and operated at the event by re:3D).
- Site: hangar at McMillan Airfield, Camp Roberts (NPS Field Laboratory), near Paso Robles, CA; grid-connected 120 VAC.
- Window: 2026-05-12 16:39 PDT to 2026-05-13 17:25 PDT (~24.8 h), spanning two distinct printing sessions.
- Engineering review: re:3D’s engineering team reviewed the findings and approved this release. Their review corrected two initial inferences, both reflected throughout: the hopper and barrel heater is DC-driven, not AC; and the current channel was mis-scaled by an internal differential burden pairing, since bench-calibrated.