Antminer – PSU Voltage Out of Range

Hard power-cycle at the breaker for a full 30 seconds, not a soft reboot. Clears any wedged PIC / `bmminer` state that can mask or fake a voltage range error and resets the PSU internal undervoltage-lockout latch. Verify the miner is completely dark at the front panel before re-energizing. If `ERR_VOLT_RANGE` clears after a breaker cycle and doesn't return for 24 hours, the warning was most likely a transient mains or boot-time event rather than hardware failure — log it, keep watching, and continue to Step 10 for longer-term mains logging if it recurs.

Reset to stock firmware profile. Strip any overclocking, undervolting, autotuning, or custom-frequency profile. Run the miner at Bitmain stock for 15 minutes and watch the log for `ERR_VOLT_RANGE` recurrence. If the error disappears under stock but returns under any OC, your profile is past the silicon-lottery ceiling for the current state of the PSU — the rail margin is too thin for aggressive tuning, and you should either back off or move to Tier 3 firmware (DCENT_OS) where per-domain tuning is visible.

Move the miner to a dedicated circuit matched to the PSU: `240 V / 15 A` (NEMA 6-15) for APW3++/APW7, `240 V / 20 A` (L6-20) for APW9/APW12, `240 V / 30 A` (L6-30) for APW17 / S21-class. One miner per circuit, no daisy-chained outlets, no undersized extension cords. This single change clears a meaningful fraction of `ERR_VOLT_RANGE` tickets at the D-Central bench. `120 V` host is tolerated only on APW3-family and only at roughly half output; anything else requires `200-240 V`.

Declutter the AC cord path between the wall and the PSU. Remove any C13/C19 adapters, pigtails, splitters, or generator-cable reducers. Factory-style direct connection only — C19 receptacle to C20 inlet on APW3++/APW7, L6-20 or L6-30 factory cord on APW9/APW12/APW17. Inspect the jumper cable for blackened pins, melted jacket, discolouration on the plug shell, or crushed strain relief; replace immediately if in doubt. Cheap adapters introduce voltage loss, poor contact, and hidden fire risk.

Check ambient temperature at the PSU intake with an IR thermometer or a stuck-on thermistor. Target `≤ 30 °C`, worst-case `≤ 40 °C`. Every 10 °C above the cap datasheet temperature roughly halves electrolytic life (Arrhenius / `10 °C rule`). An APW in a 45 °C garage ages about 4× faster than one in a 25 °C basement, and today's `ERR_VOLT_RANGE` is often last summer's heat catching up. Open a window, add an intake fan, duct cooler air, or relocate before continuing.

Meter the rail at the PSU output screws under load. DC volts, black probe on `0 V` screw, red on `12 V` / `15 V` output screw. Miner at full hashrate, hold probes steady for 60 s, log the lowest reading. Match against the rail window for your PSU family (APW3++/APW7: `12.4-13.6 V`; APW9: `12.6-14.5 V`; APW12: `12.9-15.0 V`; APW17: `14.0-16.5 V`). Outside the window under sustained load = foundational symptom. Every later step builds on this — do not skip, do not trust the filtered UI number.

Meter the rail at the hashboard input connector under the same load. If the delta between PSU output and hashboard input exceeds `0.3 V`, the jumper or voltage-regulating sense cord is dropping voltage under current — cable loss, not PSU loss. Replace the C13/C19 jumper with a fresh heavy-gauge unit, replace the sense cord with a new part of the correct PSU family, and re-measure to confirm the delta closes. Blackened pins or spread receptacles are the smoking gun; inspect both ends of every cable.

Swap in a known-good PSU of the correct family. Borrow from a friend's identical miner, buy a salvage-grade APW from D-Central's power supply inventory, or pull a known-good brick from a retired unit. Match the PSU to the miner exactly — an APW3++ belongs on an S9-class load, an APW12 on an S19, an APW17 on an S21; mismatched pairings have their own `ERR_VOLT_RANGE` signatures. Repeat Step 6. If the rail recovers on the substitute, the original PSU is degraded and needs Tier 3 cap refresh or outright replacement.

Isolate the hashboard load one board at a time. With a confirmed good PSU, disconnect two of the three (or four, on S21 Pro) hashboards and run the miner with only one board loaded. Document the rail under load for each board individually. Any board that drags the rail out of range on its own has a voltage-domain short, failed PMIC, or dead ASIC — move that board to the Tier 3 queue. A board that holds the rail cleanly when the others can't is healthy; the fault is in one of the other two or three.

Log mains voltage for 24-48 hours. Kill-A-Watt EZ, Shelly EM, or an Emporia Vue whole-panel monitor on the miner's outlet or branch circuit breaker. Anything sagging below `230 V` on split-phase or below `115 V` on a tolerated-`120 V` install points at the circuit, the service panel, or the utility — not the miner. Evening residential peak (6-10 PM) is the most common pattern; if the sag persists across seasons, escalate to an electrician for a panel / service-drop audit before spending a dollar on PSU parts.

Flash DCENT_OS for per-hashboard, per-voltage-domain visibility. DCENT_OS is D-Central's own open-source Antminer firmware — the Mining Hacker default across the Antminer lineup, open-source on GitHub, no licensing overhead. Alternatives: Braiins OS+, LuxOS, Vnish. Run 20 minutes to stabilize, record worst-reading domains. Any domain sagging `0.2 V+` below its siblings under the same load has a failing regulator or PMIC on that specific board — you've just turned a mystery `ERR_VOLT_RANGE` into a named component. Canonical: `https://d-central.tech/dcent-os/`.

Open the APW for a bulk-cap audit (Tier 3 only — dangerous). Unplug, wait 5 minutes, then bleed the primary bulk caps through a `10 kΩ / 10 W` resistor to `0 V` reference — these retain `400 V+` for minutes after unplug and will put you on the floor if shorted with a screwdriver. Visual pass: bulging tops, crusted rubber plug, brown electrolyte residue = cap is gone. ESR-meter every primary electrolytic with a Peak Atlas ESR70 or Blue ESR; anything above 3× datasheet ESR is drifted and should be replaced. Secondary-side caps age too, particularly on APW9/APW12.

Replace failed caps with 105 °C / 10,000 h-rated parts. Match voltage rating (typically `450 V` on primary bulk, `25-35 V` on secondary output), capacitance within ±10%, and physical case size. Nichicon, Rubycon, Panasonic, or United Chemi-Con low-ESR electrolytics only — avoid unknown Asian rebranders, which is how a refreshed APW ends up back on the bench in 18 months. Maintain creepage and clearance for `450 V` primary-side voltages; no shortcuts. Flux-clean all joints, inspect for cold solder, and reflow anything suspect before closing up.

Inspect and replace the PFC MOSFET and boost diode if indicated. `RDS(on)` drift is hard to measure without bench equipment, but a visibly discoloured TO-247, a cracked solder joint on the MOSFET source lead, or measurably high gate-source leakage are grounds for replacement. Use genuine `IRFP460` or equivalent `N-channel 500 V` parts from Digi-Key or Mouser on APW3-family; higher-current parts on APW9/APW12/APW17 — match the original part number exactly, no AliExpress counterfeits. Re-paste the heatsink with Arctic MX-6 and a fresh mica insulator.

Post-repair load test the APW before putting it back on a production miner. Dummy-load at the miner's nameplate power (`~1,400 W` S9; `~2,100 W` S17; `~3,250 W` S19; `~3,500 W+` S21) on a resistor bank or a retired hashboard wired as a load. Hold full load 30 minutes. Rail stable within the PSU-family window, case under `55 °C` at the vent, fan RPM appropriate = safe to reinstall. Never skip this — a PSU that looks fine at idle can still fail under sustained load, and re-introducing a marginal PSU restarts the whole failure cycle.

Ship to D-Central Tier 4 when DIY has reached its limit. Stop-and-ship triggers: visible cap leak, burn marks, or burnt-PCB smell on the APW; a substitute PSU of the correct model on a confirmed `240 V` circuit also throws `ERR_VOLT_RANGE`; per-domain voltage on DCENT_OS isolates a regulator you can't replace; you've refreshed caps once and the error returned within 90 days; you aren't set up to safely discharge a `400 V+` primary bulk cap. Book at `https://d-central.tech/services/asic-repair/`. Canadian parts, Canadian workshop, real bench equipment.

D-Central bench process. APW test fixture with a programmable AC source — we can replay the exact mains condition that killed your rail (sag, spike, `120 V` feed, brownout cycle) to confirm failure mode rather than guessing. Full component-level audit of primary bulk electrolytics, PFC MOSFET, boost diode, secondary rectifiers, control IC, and the voltage-regulating sense circuit. Parts sourced in Canada from Digi-Key / Mouser / Nichicon authorized channels — no counterfeits. 24-hour nameplate-load burn-in at room temperature before release. Hashboard per-domain PMIC replacement on the same ticket if damage propagated.

Ship safely to D-Central. Anti-static bag each hashboard individually; bag the PSU separately in an ESD-safe wrap. Double-box with ≥ 5 cm foam on every side. Tape the AC inlet shut to protect the C20 / L6 pins in transit. Include a written note: observed rail voltage idle and under load, firmware version, mains voltage at the outlet, approximate hours of operation since purchase, the specific `ERR_VOLT_RANGE` log excerpt, and your contact details. Better notes = less bench diagnostic time = lower repair cost. Canada, US, international — D-Central accepts all three.