/Documentation
Test Report

Low-Power Regen 100 A Phase-Current Endurance (No Cooling)

Twenty-two-minute continuous regenerative braking at a 100 A q-axis current step on a dynamometer-spun machine, run with the bare C2 power stage (no heatsink, no fan) to prove phase-current endurance and capture thermals.

Supported by
Mouser ElectronicsMitsubishi ElectricSendCutSend

Document Information

Document IDOV-TEST-HW-REGEN-100A-NO-COOLING
Test ID12
Version0.2
Date2026-09-22
Applies toopenvvvf-control-module, chassis-size-2
Product lineopenvvvf
Normative referencesOV-TEST-HW-INDEX
Index / Safety and Compliance / Testing and Validation / Hardware Tests / Low-Power Regen 100 A Phase-Current Endurance (No Cooling) / Low-Power Regen 100 A Phase-Current Endurance (No Cooling)
12 · v0.2 · 2026-09-22Download PDF
DoctypeTest Report
Product lineopenvvvf
Applies toopenvvvf-control-module, chassis-size-2
Version0.2
Date2026-09-22
DescriptionTwenty-two-minute continuous regenerative braking at a 100 A q-axis current step on a dynamometer-spun machine, run with the bare C2 power stage (no heatsink, no fan) to prove phase-current endurance and capture thermals.
Test id12
Nav order359
Normative refsOV-TEST-HW-INDEX

Low-Power Regen 100 A Phase-Current Endurance (No Cooling)

This report records a regenerative-braking endurance run on the OpenVVVF C2-class power stage. A Sierra CP Engineering dynamometer spun the coupled machine at ~280 RPM while the inverter applied negative q-axis current in 10 A steps up to −100 A, then held −100 A for 22 minutes. The point of the test is in the name: the DC bus was only ~52 V, so 100 A of phase current amounts to roughly 300 W of electrical power — full phase-current stress at very low power — and the bare board stack ran with no heatsink and no forced-air cooling the entire time.

Test setup

  • DUT: OpenVVVF control module on a C2-class power stage with the dynamometer termination/connection board (silkscreen Rev 2). The stack is bare on an aluminum plate: no heatsink, no fan, natural convection only.
  • Machine: 10-pole (5 pole-pair) 3-phase induction machine, shaft driven by the dynamometer through a torque flange (ATO-101-005) at ~280 RPM.
  • DC supply: BK Precision bench supplies holding a ~52 VDC bus; regenerated energy returned to the bus/supply.
  • Instrumentation: RTE Studio telemetry logged at ~14 Hz (19,463 samples), Klein Tools TI250 thermal imager, Tektronix 5 Series MSO, Sierra Trakkor dyno software.
DUT close-up: bare OpenVVVF power-stage stack on an aluminum plate, no heatsink or fan
DUT close-up: bare OpenVVVF power-stage stack on an aluminum plate, no heatsink or fan
Bench: DUT with BK Precision supplies, scope, and thermal imager
Bench: DUT with BK Precision supplies, scope, and thermal imager
Test cell overview with the dynamometer and ABB cabinet (80 V marker on wall)
Test cell overview with the dynamometer and ABB cabinet (80 V marker on wall)
Dynamometer coupling: flywheel, torque flange ATO-101-005, and the driven machine
Dynamometer coupling: flywheel, torque flange ATO-101-005, and the driven machine
Trakkor dyno screen during the run: ~280 RPM shaft speed
Trakkor dyno screen during the run: ~280 RPM shaft speed

Test conditions

Parameter Value
DC bus voltage ~52 VDC (50.6–52.6 V observed)
Shaft speed ~280 RPM (dynamometer-held; sagged to ~263 RPM at full regen torque)
Current command Direct IqVar override, 0 → −10 → … → −100 A in 10 A steps, then hold
Hold at −100 A 22.0 min (t = 35 s to 1355 s of the log)
Electrical power at 100 A ~296 W average into the DC bus (~5.8 A DC-link)
Ambient temperature ~23–27 °C
Cooling None (open-air natural convection, bare stack)
Throttle inputs 0 throughout (pure current override)

Procedure

  1. Dynamometer spins the machine at ~286 RPM with the DUT enabled and IqVar = 0 (baseline, ~9 s).
  2. IqVar stepped −10 A per command to −100 A over ~25 s.
  3. Held at −100 A for 22 minutes while telemetry and thermal images were captured.
  4. Released to IqVar = 0; logging continued ~20 s into cooldown.

Electrical results

Per-step statistics from the full-rate telemetry log (Pdc = DC-link power, Idc = DC-link current):

Iq command Step duration Avg Iq Peak |phase| Avg Pdc Avg Idc
0 (baseline) 9.3 s +0.1 A 4 A −30 W (supply feed) −0.6 A
−10 A 4.2 s −9.9 A 13 A 23 W 0.4 A
−20 A 2.0 s −19.0 A 21 A 69 W 1.3 A
−30 A 3.7 s −32.8 A 50 A 108 W 2.1 A
−40 A 2.0 s −35.3 A 48 A 149 W 2.9 A
−50 A 3.3 s −50.0 A 64 A 171 W 3.3 A
−60 A 2.1 s −55.3 A 74 A 200 W 3.9 A
−70 A 2.4 s −68.7 A 86 A 238 W 4.6 A
−80 A 2.4 s −81.6 A 104 A 264 W 5.1 A
−90 A 3.4 s −90.3 A 118 A 279 W 5.4 A
−100 A (hold) 22.0 min −100.0 A 145 A 299 W 5.8 A

During the 22-minute hold:

  • Average q-axis current was −100.0 A with every one of the 16,628 settled samples below zero — continuous, uninterrupted regen.
  • Instantaneous phase-current peaks reached ±145 A (100 A fundamental plus switching ripple and brief id excursions).
  • DC-link power averaged 299 W (max 381 W) at ~5.8 A; total regen energy over the hold was ≈390 kJ (≈108 Wh), ≈109 Wh for the whole run. This is the "low power" in the title: full phase current, but only ~0.3 kW because the bus is 52 V.
  • Bus voltage stayed in a tight 50.6–52.6 V band with no sign of regeneration-induced overvoltage.

Thermal results

Board NTCs (inside the stack) climbed steadily and plateaued, with no runaway:

  • Inverter temp 1: 33.4 °C → 79.1 °C (peak 79.4 °C) — about +46 K
  • Inverter temp 2: 35.6 °C → 80.3 °C (peak 80.4 °C, reached seconds after release) — about +45 K
  • After release, both sensors began cooling immediately (77.7 / 79.1 °C by the end of the log).

Thermal imager spot readings on the exterior (all taken during the −100 A hold):

Board, early in the hold (~7 min)

TI250: power board ~7 min into the hold, 28.5 °C at the crosshair
TI250: power board ~7 min into the hold, 28.5 °C at the crosshair
  • Crosshair: 28.5 °C; imager span 21.9–41.4 °C.
TI250: DC-link capacitors ~7 min into the hold, 49.4 °C at the crosshair, 52.7 °C max
TI250: DC-link capacitors ~7 min into the hold, 49.4 °C at the crosshair, 52.7 °C max
  • Crosshair: 49.4 °C; max 52.7 °C.
TI250: DC-link capacitors ~8 min into the hold, 52.1 °C at the crosshair, 56.3 °C max
TI250: DC-link capacitors ~8 min into the hold, 52.1 °C at the crosshair, 56.3 °C max
  • Crosshair: 52.1 °C; max 56.3 °C.

Board, late in the hold (~18 min)

TI250: power board ~18 min into the hold, 37.6 °C at the crosshair, 50.1 °C max
TI250: power board ~18 min into the hold, 37.6 °C at the crosshair, 50.1 °C max
  • Crosshair: 37.6 °C; max 50.1 °C.

Phase cabling / lugs, late (~18 min)

TI250: phase cabling and lugs ~18 min into the hold, 71.5 °C at the crosshair — hottest exterior point
TI250: phase cabling and lugs ~18 min into the hold, 71.5 °C at the crosshair — hottest exterior point
  • Crosshair: 71.5 °C; max 70.7 °C scale top — the hottest exterior surface observed.
TI250: DC-link capacitors ~18 min into the hold, 40.4 °C at the crosshair, 64.3 °C max
TI250: DC-link capacitors ~18 min into the hold, 40.4 °C at the crosshair, 64.3 °C max
  • Crosshair: 40.4 °C; max 64.3 °C along the capacitor bodies.

End state on the RTE Studio screen (~19 min)

RTE Studio at the end of the hold: IqVar command history, temps ~76–77 °C, 268 W DC-link
RTE Studio at the end of the hold: IqVar command history, temps ~76–77 °C, 268 W DC-link
  • Console history shows the final steps down to IqVar = -100; live values read 76.3 / 77.1 °C board temps with 5.16 A / 268 W on the DC link — consistent with the logged data.

Telemetry overview

Telemetry overview of the full run: phase currents, DQ currents, DC-link power, and board temperatures
Telemetry overview of the full run: phase currents, DQ currents, DC-link power, and board temperatures

Open telemetry log: View the run in the Telemetry Viewer

Observations

  • No faults, trips, or resets at any point; the current loop held −100 A for the full 22 minutes without intervention.
  • This is deliberately a current stress, not a power stress: at 52 VDC the bus sees only ~6 A even at 100 A phase current. It verifies conduction paths, current sensing, and gate drive at rated phase current with essentially no cooling infrastructure.
  • Machine-terminal power is not quoted for this run: P = (3/2)(vd·id + vq·iq) from this session's logged dq voltages does not reconcile with DC-link power — the implied ≈0.7 kW of inverter dissipation is inconsistent with a bare stack plateauing at 80.4 °C. The dq-voltage calculation reconciles in firmware build 855dbcee90d611cb and is reported from the 400 A run onward.
  • Shaft speed sagged from ~280 to ~263 RPM as regen torque loaded the dynamometer's speed control.
  • The temp_inv3_c and temp_motor_c channels were unpopulated (null) for this build.
  • Interior NTCs read ~10–15 °C above the hottest exterior imager spots — expected, as the sensors sit inside the stack next to the dissipating devices.
  • Thermal imager readings are crosshair/spot values from an auto-scaled palette; treat them as ±2–3 °C.

Conclusion

Pass. The bare C2-class power stage sustained a 100 A regenerative phase current for 22 minutes with no cooling and no faults. Peak board temperature was 80.4 °C (≈+46 K over ambient start) with a clear plateau, and the hottest exterior point (phase lugs) reached ~71.5 °C. The power stage demonstrably tolerates rated phase current at low DC-bus voltage on natural convection alone, with comfortable thermal margin. This run complements the no-load and 180 V induction-motor bring-up tests by covering the high-current/low-voltage regen corner.

Artifacts

Notes

  • The full-rate log (19,463 samples, 16 MB) is retained alongside this report as 100a-regen-test.jsonl (not committed); the linked artifact is decimated to one sample per second for the repository.
  • Photo elapsed times assume the camera clock (UTC-7) matches the session wall clock; the telemetry session ran 08:26–08:49 UTC on 2026-09-22.
  • The dyno screen photo shows the cell's Trakkor display (shaft speed ~280 RPM) for context; electrical values in this report are from the DUT's own telemetry unless stated.