ASIC 220V vs 110V Power Derating Explained
Informational — Monitor and address as needed
Symptoms
- PSU label reads `100-240 V AC` or `200-240 V` and the outlet feeding it is `120 V` (NEMA 5-15 / 5-20)
- Realized hashrate sits 15-30% below nameplate with all chains alive and HW% in spec
- Wall ammeter reads 13-15 A continuous on a 15 A 120 V circuit; breaker warm to the touch
- PSU fan running at 100% duty cycle within 60 s of full load — input rectifier and PFC stage thermally stressed
- Audible 120 Hz hum or low-frequency buzz from the PSU brick at low input voltage
- Power factor / efficiency reading on a Kill-A-Watt or Emporia Vue is 87-89% instead of the spec'd 93-94%
- PSU refuses to start or shows red LED — late-rev APW12, APW17 / APW171, P221 will not initialize below ~205 V
- Voltage logger shows mains sagging from 120 V idle to 108-114 V under load on standard #14 AWG wiring
- Cord, plug, or PSU inlet warm-to-hot after 30 min of mining (heat = current² × resistance, current is doubled at 110 V)
- Circuit breaker trips intermittently 5-15 min into a hashing session
- Tried everything on the miner side (firmware, OC profiles, hashboard swaps) and the hashrate ceiling never moves
- kern.log / cgminer reports `ERROR_POWER_LOST` or `power voltage can not meet the target` on a PSU that label-tested fine
- Operating in Canada or northern US with a 240 V dryer outlet 15 ft away that you've never been told you could use
Step-by-Step Fix
Verify what you have at the wall. Walk to the panel, find the breaker for the mining outlet — single-pole = 120 V, double-pole or ganged = 240 V split-phase. Then walk to the receptacle and identify the NEMA designation (5-15/5-20 = 120 V; 6-15/6-20/L6-20/L6-30/14-30/14-50 = 240 V). Trust the breaker, not the receptacle; a 240 V-style outlet wired to a 120 V breaker is a common install error documented across NEC inspections.
Multimeter the outlet on AC volts. On 240 V receptacles probe hot-to-hot; on 120 V probe hot-to-neutral. Expect 235-245 V on 240 V split-phase residential, 202-212 V on 208 V three-phase commercial, 118-124 V on 120 V household. If you read 115-125 V across the two hot pins of what should be a 240 V outlet, the outlet is mis-wired or actually single-phase 120 V. Do not plug an ASIC into it.
Check the PSU label and pull the full datasheet. Look for `Maximum Input Current` and any second output rating for 100-120 V input. Bitmain, MicroBT, and Canaan all publish 100-120 V derate output numbers somewhere — Bitmain on its support portal, MicroBT in the P21/P221 manual PDFs, Canaan in the AvalonMiner manual. Photo the label for your records and the eventual repair ticket.
Calculate your real wattage budget: Max input amps × actual outlet voltage × 0.92 efficiency = honest output watts. Compare to the miner's nameplate. If the gap is more than ~10%, the PSU is undersized for the current voltage — you need a 240 V circuit, a smaller miner, or a different PSU. Common fleet derates: APW3++ ~20%, APW7 ~22%, APW9 ~33%, early APW12 ~30%, APW17/171 hard-stop.
Move to an existing 240 V circuit if you have one. Most North American homes already have 240 V at the dryer (NEMA 14-30), electric range (NEMA 14-50), EV charger, well pump, or central A/C disconnect. Adapters from 14-30 / 14-50 to L6-20 or 6-20 are inexpensive. Do not run a miner permanently from a dryer outlet without a transfer switch — that's a code issue — but it's a valid temporary feed for testing.
Install a dedicated 240 V circuit. Hire a licensed electrician or pull a permit yourself where allowed. Spec to match the miner: S9/S9 SE/L3+/Bitaxe rack = 15 A at 240 V (NEMA 6-15); S17/S19/S19 Pro/S19 XP class = 20 A at 240 V (NEMA 6-20 or L6-20, 12 AWG); S21/S21 Pro/hydro = 30 A at 240 V (L6-30, 10 AWG); multi-rig farm = 50 A at 240 V (14-50). Dedicated means the miner is the only load on that breaker.
Log voltage under load for 24-48 hours with a Shelly EM, Emporia Vue, or Sense energy monitor. Acceptable on 240 V split-phase: 225-250 V. On 208 V three-phase: 195-218 V. Sags outside those ranges = panel feed undersized or utility transformer loaded — that's a panel or service issue, not a miner issue, and it'll fight you on every rig you add.
Verify miner-to-PSU pairing against the manufacturer matrix. Bitmain's PSU compatibility table at support.bitmain.com/hc/en-us/articles/15909673875993 is canonical. Common pitfalls: APW3++ on S19 (caps at ~1800 W, miner needs 3250 W); APW9 on S21 (trips OCP under load); APW12-12/15 on S21 (wrong output voltage window — S21 needs APW17 / APW171215a). Wrong PSU model + correct input voltage still throws PSU_ERR.
Use a Kill-A-Watt EZ P4460 or equivalent for spot voltage / amps / power-factor checks. A $40 plug-through monitor catches 90% of voltage-class problems in 30 seconds. Plug it between the wall and the PSU. Read AC volts (input voltage), amps (input current), and PF. PF below 0.92 on a switching PSU = something is off. Volts below the PSU's spec'd minimum input = derate territory and time to upgrade the circuit.
Install whole-circuit surge protection. Type-2 SPD at the panel + Type-3 at the miner outlet. Mis-voltage events often pair with other transients (lightning, utility switching, neighbour's big motor cycling). A $150 panel SPD plus a $30 outlet SPD pays for itself the first time it catches a transient that would otherwise toast an APW17. Eaton CHSPT2ULTRA is a good Canadian residential pick.
Audit the input-side electrolytics if the PSU has been on 110 V for 30+ days. Operating at the bottom of the input voltage range accelerates aging on primary bulk caps and the PFC MOSFET. Bleed bulk caps through a 10 kΩ / 10 W resistor to ground (they hold 400+ V for minutes after unplug). ESR-meter the primary electrolytics — any cap above ~3× rated ESR is drifted; bulged or leaking caps are dead. Common APW casualty list: primary bulk on the HV rail, PFC MOSFET, occasionally the high-side gate-drive opto.
Replace failed components with proper parts. Bulk caps: 450 V rating on APW primary, match capacitance and life grade (105 °C, 10,000 h minimum). PFC MOSFET: genuine part from a reputable distributor — counterfeit MOSFETs are how you end up back in this fix in six months. Re-paste any thermal interfaces with proper paste plus mica. Reflow only with a hot-air station and a temperature profile you trust.
Post-repair load test before reinstalling. Dummy-load the repaired PSU at nameplate wattage (resistor bank or a retired hashboard wired as test load). Watch rail voltage and PSU enclosure temp for 30 min. Rail stable within spec, case under 55 °C at vent, fan RPM appropriate = safe to reinstall. Never plug a bench-repaired PSU back onto a production miner without a load test — the whole point of Tier 3 is not repeating the failure in production.
Hashboard audit after extended low-voltage operation. If the PSU was sagging its 12/15 V output rail under load (which 110 V operation forces it to), the hashboards saw rail dips. On Antminer hardware: flash DCENT_OS for per-chip HW% and per-domain voltage. Run a 20 min soak; any chip position consistently elevated >5% HW% = voltage-domain stress damage. On Whatsminer use MinerTool diagnostics; on Avalon use the AvalonMiner dashboard. DCENT_OS coverage for those platforms is on the roadmap, not yet shipping.
Document the incident and PSU history. Photos of the original install, the PSU label, the kern.log / cgminer log lines, receipts for any damage. If a reseller shipped you the wrong cord, that's their liability. If an electrician mis-wired the receptacle, that's theirs. Insurance, chargeback, and Small Claims all need the paper trail; D-Central has had customers recover thousands of dollars this way.
Stop DIY when any of these are true: (a) PSU shows visible cap bulging, leaking, or burn marks; (b) electronics / ozone / hot-PCB smell after the incident; (c) substitute PSU of the correct model also fails on the now-confirmed 240 V circuit; (d) hashboard HW% spikes after 30+ days of 110 V operation; (e) you measured 400+ V on a primary bulk cap and aren't set up to bleed it safely. Book a D-Central repair slot at d-central.tech/services/asic-repair/.
What D-Central does at the bench: programmable AC source can simulate the exact mis-voltage event that aged your PSU to confirm failure mode, full component-level audit (caps, MOSFETs, diodes, optos, control IC), Canadian-sourced replacement parts (no AliExpress counterfeits), 24 h nameplate-load burn-in before return. If hashboards also took a hit, we diagnose and repair on the same ticket — voltage-domain PMICs and per-domain input filter caps are the usual suspects.
Ship safely. Disconnect PSU from miner. Tape AC inlet shut to protect pins. PSU in anti-static bag, double-boxed with ≥5 cm foam on every side. Hashboards (if also being sent): anti-static bags, double-box separately. Include a typed note: observed mis-voltage event, measured mains voltage, time on the wrong feed, symptoms after move to correct voltage, your contact info. That note saves bench time, which saves you money.
When to Seek Professional Repair
If the steps above do not resolve the issue, or if you are not comfortable performing these repairs yourself, professional service is recommended. Attempting advanced repairs without proper equipment can cause further damage.
Related Error Codes
Still Having Issues?
Our team of Bitcoin Mining Hackers has been repairing ASIC miners since 2016. We have seen it all and fixed it all. Get a professional diagnosis.
