Whatsminer Error 270 – Power Input Overvoltage Protection

DO NOT re-energise the miner without measuring outlet voltage first. The single biggest mistake on a Code 270 event is treating it like an output-side fault and just power-cycling. If the input overvoltage condition is still present, the next power-up makes the damage permanent. DMM on AC volts, probe the outlet with the miner unplugged, confirm 235-245V L-L on a 240V service or 200-212V L-L on a 208V service before plugging anything back in. Document the reading. If outside spec, stop and escalate to Tier 2 / electrician.

Power-cycle at the AC mains, not the chassis switch. Once outlet voltage is confirmed in spec, kill AC at the PDU or wall breaker, wait a full 60 seconds for primary-side bulk caps and secondary-side output caps to discharge. A short chassis-button reboot leaves the PSU MCU's fault latch set on some firmware revisions and Code 270 re-appears instantly. A full AC cycle resets the latch cleanly. If the error returns on power-up with confirmed-good input voltage, the PSU has internal damage from the prior event - escalate to Tier 3 / Tier 4.

Cross-reference the fault timestamp with local electrical events. Pull log/btminer.log and find the exact Code 270 timestamp. Check blitzortung.org for lightning within ~3 km in the prior 10 minutes. Check the local utility outage map. Did anyone in the household run a welder, EV charger, well pump, or other large motor load just before the trip? Was a generator running? Was an electrician on-site in the prior 48h? Identifying the trigger is half the fix; without it, you are guaranteed a repeat event.

Visually inspect the cord, plug, and outlet. Discoloured or burnt plug pins indicate a high-resistance connection (warm under load), which can cause voltage transients on disconnect events. Loose outlet receptacles (cord can wiggle in the socket) are a slower-but-real cause of erratic input voltage. Replace any cord with damaged insulation or scorched plug pins. Replace any outlet that does not grip the plug firmly. Cheap fix, high-yield.

Verify the cord matches the outlet matches the panel voltage. A NEMA 6-20P cord plugs into both 240V residential and 208V Wye commercial outlets - physically. The cord does not know what is on the other side. Same for L6-30 locking plugs. Always meter the actual outlet voltage with a DMM before plugging in a Whatsminer. If you find a 277V outlet on what you thought was a 240V service, you have found a major hazard requiring electrician intervention - do not power the miner.

Log mains voltage continuously for 24-72 hours with a power monitor. A handheld DMM only catches steady-state. The transient that triggered Code 270 may have been a <100 ms spike that will not recur in the time you stand there with a meter. A true-RMS power logger (Fluke 1736, Hioki PW3360, Owon PDS5022, or an Emporia Vue / Sense energy monitor for budget builds) logs at high sample rate and captures the events you need to see. Under CAD $200 for a basic logger; $1000+ for a Fluke. Plot the log and look for the spike correlating with your trip timestamp.

Install a Type-2 whole-panel surge protective device (SPD). This is the single most cost-effective fix for Code 270 events caused by transients: utility switching, induced lightning, large neighbourhood load drops. Models like the Intermatic IG2240-IMSK, Siemens FS140, or Square D HEPD80 install at the panel in 30-60 minutes of electrician time and cost CAD $150-300 plus labour. Type-2 SPD spec: 40 kA per phase minimum, 1.5 kV clamping voltage or lower, indicator light to confirm the device is still alive after a hit. Replace the SPD after any major event - they wear out, and a dead SPD looks identical to a live one until you check the indicator.

Add a series-mode surge filter or voltage stabiliser for high-risk circuits. On generator-fed sites, on services with documented poor power quality, or downstream of a Type-2 SPD where extra protection is wanted: a series-mode filter (SurgeX class) or a series voltage stabiliser (APC Line-R class for small loads, or a dedicated industrial unit for >3 kW Whatsminer loads) provides another 1-2 kV of clamping and corrects steady-state voltage drift. Cost: CAD $200-1,500 depending on rating. Size at 1.5 x peak miner draw. Worth it for off-grid / generator deployments and for any site with multiple Code 270 events.

For generator-fed sites: oversize the genset and verify AVR settings. A 30 kW genset feeding a 5.4 kW M50S++ at ~18% of rated load handles miner switching events cleanly. The same M50S++ on a 7.5 kW genset (72% of rated) is asking for Code 270 overshoots every time the miner cycles tune or another load on the genset switches. Rule of thumb: total miner load <30% of genset rated output, with the rest of the headroom for switching transients. Have a qualified genset tech verify AVR setpoint, gain, and droop are tuned for ASIC duty cycle (<1% voltage overshoot on a 25% load step is the target).

Replace the cord and the outlet receptacle even if they look fine. After a Code 270 event with confirmed transient cause, the miner cord and the outlet have both absorbed energy. Insulation degradation and contact pitting from a single major event are not always visible but absolutely affect the next event. New cord (CAD $30-80 for an OEM Whatsminer or quality replacement) and new commercial-grade outlet (CAD $20-40 for a Hubbell or P&S spec-grade 30A 240V receptacle) is 1-2 hours of work that pays for itself the next time a transient hits.

Verify the panel ground / neutral bond at the service entrance. On a 120/240V split-phase service, the neutral is bonded to ground exactly once - at the main service disconnect. A loose or missing bond is the textbook cause of lost-neutral overvoltage events. With main breaker OFF, visually verify the neutral busbar has a heavy copper conductor (4-6 AWG minimum) bonded to the ground busbar and to the grounding electrode. Also verify any panel-mounted grounding electrode conductor is intact. This is electrician work; if you do not already know what these terms mean from your panel, hire one.

Pull the PSU and visually inspect the input section under magnification. Kill AC, wait 5 minutes, remove the PSU from the chassis, open the PSU-side cover. Under bright light and 10x magnification, inspect the input MOV disc (~10-14 mm blue/orange epoxy near the AC inlet) for discolouration / cracks / scorch; the input bulk capacitor (~470 uF, 450V) for bulged top, vent leak, or electrolyte residue; the PFC MOSFET package(s) for scorch or lifted leg; the AC inlet fuse for an open or thermally-stressed body; the input-side PCB tracks for brown / yellow heat marks. Any single visible damage = PSU has lost its margin even if it still functions; replace. Multiple visible damage = unit is dangerous; do not power, ship to D-Central or recycle.

Replace the input MOV array (skill gate). If the only visible damage is the MOV disc, on some P21 / P221 revisions the MOV is field-replaceable: heat the through-hole pads with a 60W iron + flux, lift the old disc, install a matching 14D471K or 14D561K (verify part number from silkscreen and against the original - ~471V vs ~561V clamp varies by PSU revision). Replace ALL input-side MOVs, not just the visibly-damaged one - adjacent MOVs share clamp duty and have absorbed similar cumulative joules. Cost: CAD $5-15 in parts, 30 minutes of work for a competent solderer. Test with bench AC source if available; otherwise re-install and monitor.

Replace the input bulk capacitor. A bulged or vented bulk cap is unconditionally end-of-life. Match capacitance, voltage, and ripple-current rating from the silkscreen - typical spec on a Whatsminer P21 input is 470 uF, 450V, 105C, low-ESR, ripple current >2 A. Quality brands: Nichicon, Rubycon, Panasonic, KEMET. Avoid no-name parts on this rail - counterfeit caps fail catastrophically under PFC stress. CAD $8-20 per cap, 30 minutes for a confident through-hole solderer. After a bulk-cap replacement, the PSU should be bench-tested before re-installation; the rest of the input section may also be stressed.

Commission a generator AVR for ASIC duty (off-grid only, qualified genset tech). AVR commissioning involves: voltage setpoint calibration to nominal 240V or 208V, gain tuning for fast settling (<200 ms response to a 25% load step), droop tuning for parallel-set stability if applicable, and verification with a recording oscilloscope at the genset terminals. This is 2-4 hours of skilled work that converts a miner-hostile generator into one that will not trip Code 270 on every load step. Pair with a stabiliser downstream for belt-and-suspenders protection.

Install a sub-panel with dedicated SPD for the mining circuit. On large home setups (3+ Whatsminers) or container farms, a dedicated mining sub-panel with its own Type-2 SPD upstream of the mining branch circuits provides defense-in-depth protection. Sub-panel install is CAD $400-1,200 depending on local pricing and complexity, plus the SPD. Provides electrical isolation between mining loads and household / office loads, simplifies metering, and lets you kill mining circuits without touching the main panel during service.

Investigate a stuck or flapping neutral with the utility (service-call only). If Step 4 / Step 7 indicate a lost or high-impedance neutral, this is utility territory. Symptoms: voltages on 120V outlets vary widely depending on household load, lights brighten when large loads turn off, multiple Code 270 events correlated with random household activity. Call the utility customer service line, request an emergency service inspection, do NOT re-energise mining loads until the utility has cleared the service. A lost neutral can kill people - it is one of the few electrical fault categories where the utility will dispatch crews after-hours.

Install a transient event recorder for diagnostic confirmation. A dedicated transient event recorder (Dranetz HDPQ, Fluke 1760, or open-source equivalent) logs voltage waveforms at sufficient sample rate to capture the actual transient shape that is tripping Code 270. Costs CAD $2,000+ for commercial gear, but rentable from electrical-test-equipment vendors for ~CAD $100/day. The waveform tells you whether it is a slow swell (utility transformer overload), a fast spike (lightning / switching), a notch-recovery overshoot (motor stop), or a generator overshoot - each has a different fix.

Stop DIY when Tiers 1-3 are clean but the PSU still trips Code 270 on confirmed-good input. This signals internal PSU degradation: the input MOVs, the bulk caps, or the PFC stage have lost margin from prior absorbed events. Bench testing with a programmable AC source is the only way to characterise the actual trip threshold and confirm replacement is needed. Home gear cannot reproduce the conditions safely. D-Central's bench process tests against a Pacific Power-class AC source at 200-300V ramp profiles, characterises the trip threshold, replaces degraded parts to OEM spec, and burns in for 24 hours at full load before return.

Stop DIY when you see scorched silicon, blued copper, vented capacitors, or visible burn marks on the input section. Cascade damage from a major surge event leaves multiple stressed-but-not-yet-failed parts adjacent to the visibly-failed parts. Replacing only the visibly-broken component and re-energising under load is how an R&R event becomes a destructive failure with hashboard collateral damage. D-Central's bench process inspects and recaps the full input-protection chain, replaces damaged MOSFETs, MOVs, and bulk caps, and validates against full-load duty cycle before return.

Ship with full context. Pack the chassis with the PSU (we need your exact stack to correlate input damage with output behaviour), a copy of the btminer-api error_codes history ([270], [270, 233], [270, 271] patterns matter), any power-logger trace from Tier 2 (CSV or screenshot is fine), service history (recent electrical work, generator events, lightning correlation, neutral-suspect events), and your steady-state input voltage measurement from Step 1. Match chassis serial to PSU serial in your ship note. Canada-wide standard shipping; US / international welcomed. Turnaround 4-8 business days for isolated Code 270, longer if cascade damage is found on the bench.

Discuss repair-vs-replace honestly. A brand-new MicroBT P21 runs CAD $220-320, P221 $340-420, P222 $420-540 from D-Central; graded-salvage units at ~60% of new. Bench-level input-section repair (MOV array + bulk cap + PFC inspection / replace) runs CAD $120-240 depending on damage extent. If the PSU is 5+ years old and has had multiple Code 270 events, repair is usually false economy - the rest of the input section is on the same aging curve. D-Central quotes up front with a photo-documented diagnostic so you can choose. ALWAYS pair the PSU repair / replacement with the upstream electrical fix - a fresh PSU on an unprotected service that produced the original Code 270 is just a fresh victim waiting for the next event.