Whatsminer Error 202 / 203 – PSU Voltage or Temperature Protection

Power-cycle at the AC mains for 60 seconds, not at the chassis switch. Kill AC at the PDU or wall breaker, wait a full minute for primary-side bulk capacitors and secondary-side output caps to discharge. A short chassis-button reboot leaves the PSU MCU's protection latch set on some firmware revisions and Code 202 re-appears instantly. A full AC cycle resets the latch cleanly. If the error clears on AC cycle and returns within minutes under load, you have a real fault, not a stuck flag. If it clears and stays clear for hours, log the event but stay alert - single events almost always recur.

Confirm Code 202 via the BTMiner API, not just the Web UI. Curl port 4028 with {"cmd":"status"} from a laptop on the same subnet and parse the error_codes array. The Web UI sometimes shows a stale fault from the previous boot; the API always shows current state. If you cannot reach the API, fall back to the LED pattern from the Whatsminer LED status reference and proceed as umbrella 202. Log the timestamp with every event so you can spot patterns later.

Check the last 24-72h for environmental, electrical, or service events. Was it hot today? Did a thunderstorm roll through last night? Did your dryer / EV charger / A/C cycle on right before the fault? Did you move the miner, swap a PSU, or open the chassis recently? Recent thermal events point at airflow or ambient. Recent electrical events point at AC infrastructure. Recent physical events point at harness or connector. Without a service log you re-diagnose from scratch every event; with one, the root cause is usually 30 seconds of review.

Inspect intake airflow and ambient with an IR thermometer. Pop the intake grille, check for dust mats, verify nothing is blocking air within 15 cm of the front. Check fan RPM (>=3,500 RPM on M30S class, >=5,500 RPM on M50S+ class). Point the IR thermometer at the intake grille - >35C means your room is hot. Check PSU exhaust under load - >85C body temperature means insufficient airflow through the PSU itself. Canadian winter mining: essentially never your problem from November to April.

Check whether the miner is on a shared circuit and back off shared loads. If the breaker feeding the miner also feeds a dryer, EV charger, A/C compressor, microwave, workshop air compressor, or kitchen circuit, every cycle of that other load is a momentary AC sag your PSU is reporting as Code 202. Easy short-term test: turn off everything else on the circuit, run the miner alone, see if 202 events stop. If yes, your circuit is undersized for the combined load - plan a dedicated circuit. If no, the issue is upstream of your panel.

Measure AC input voltage under boot load with a clamp meter at the C19 inlet. Cold-start the miner, watch the voltage during the first 60 seconds of boot and ramp-up. 240V nominal should stay >225V through inrush; 208V Wye should stay >200V. If you sag below those, your electrical feeder is undersized for the miner's peak draw and the PSU is correctly defending itself. Fix: dedicated 240V 30A circuit, #10 AWG wiring on runs >8m, no shared loads, ideally a PDU that measures and logs voltage so you catch intermittent events later. Watch for the opposite problem too - transients above 260V from generator / solar / damaged-neutral can spike the OVP path.

Capture WhatsMinerTool PSU telemetry during the next boot attempt. WhatsMinerTool exposes PSU telemetry the Web UI hides - primary-side rail voltage, internal temperature, AC input as the PSU sees it. Capture during a full boot attempt and overlay with your clamp-meter reading at the inlet. If the PSU reads AC differently than your clamp meter, the input wiring or the PSU's input sense circuit is the issue. If they agree but the primary rail is unstable, you are looking at aged caps. This 30-minute exercise saves hours of swap-and-pray downstream.

Inspect the C19 cordset and chassis AC wiring end-to-end. Power off, unplug, wait 60s. Pull the C19 cordset, inspect both ends under bright light for discolored pins, melted connector body, scorched jacket, or oxidation. Open the chassis (PSU side) to expose the AC input wiring from C19 inlet to PSU primary terminals. Look for heat-damaged jacket, loose terminal screws, signs of arcing. Wiggle each connection with the unit unplugged - anything that feels loose IS loose. DMM continuity check end-to-end on each conductor. Replace anything suspect; genuine MicroBT cordset is CAD $15-35.

Improve airflow before opening anything else. If Tier 1 confirmed >35C intake or restricted airflow, fix that physically. Pull the miner into open air, run for a day, see if fault frequency drops. Permanent fix is proper ducting (D-Central stocks universal ASIC shrouds and duct adapters), or a Bitcoin Space Heater configuration that uses the miner's heat output as home heat instead of fighting it with extra A/C. Canadian operators in October-April basically get this for free; warm-climate operators need ducting or a cooler deployment space.

Test on a known-good circuit and verify the panel itself. Run an extension cord from a different 240V circuit in your panel and plug the miner into that. If Code 202 clears, your original circuit is the problem (undersized wire, oxidized breaker terminals, shared loads). If 202 follows the miner to a different circuit, the issue is in the chassis or PSU. Also check your panel itself: terminals tight, breakers seated firmly, no discoloration around lugs. Loose neutral or main lug = whole-house voltage instability that no miner can compensate for; that is electrician territory.

Roll back any recent firmware update. If Code 202 started right after a firmware update, roll back to the previous version via the SD-card recovery procedure from WhatsMiner support. Some BTMiner builds shipped with PSU-protection-threshold tweaks that produced false 202 on certain hardware revisions. Verify firmware-to-hardware revision match before flashing - wrong firmware bricks the control board.

Measure primary-side capacitor ESR with the PSU bench-isolated. Pull the PSU from the chassis, unplug for 30 minutes (verify with DMM across each cap before touching - bulk caps hold lethal voltage longer than expected). With an ESR meter (Peak ESR70, Bob Parker, Atlas) measure each primary-side bulk cap in-circuit. Healthy: ESR within manufacturer spec, typically <0.1 Ohm for 220-470 uF / 450V bulks. Unhealthy: ESR >0.3 Ohm, bulging tops, electrolyte residue at the base. Document the measurements; this is the only objective test that distinguishes 'PSU is fine, your AC is bad' from 'your PSU caps are cooked'.

Replace failed primary-side capacitors with rated parts. If ESR is out of spec, replace the failing cap(s) with parts of equal or higher voltage rating, equal capacitance, and equal-or-better ESR / ripple-current rating. Match the temperature rating (typically 105C long-life electrolytics). Reflow with a temperature-controlled iron at 370-400C for through-hole parts; respect polarity. Parts cost CAD $20-40 typical; labor is yours. After replacement, let the unit run on a dummy load for an hour before re-installing in the chassis.

Check the AC input filter components on the PSU primary side. Less common but real: input EMI / RFI filter caps (X and Y class safety caps) crack with age and voltage stress, allowing transients to reach the active-PFC stage and trip protection. Visual inspection under magnification looks for hairline cracks in the cap bodies. These are safety-rated parts - replace only with same-class safety caps, never substitute regular ceramics. Failure mode is a drifting Code 202 that gets worse over months; replacement clears it.

Replace the PSU input-side fuse and TVS / MOV protection. If a recent surge event tripped the Code 202, the PSU's input-side surge protection (gas-discharge tubes, MOVs, TVS diodes) may have absorbed the transient and degraded. Visual: discolored MOV body, blown ceramic of a GDT, scorched silkscreen. Replace as a set - partial replacement leaves the unit vulnerable to the next event. Document the fuse rating from the silkscreen and source the exact part - substituting a higher-rated fuse is unsafe and voids any remaining warranty.

Verify primary-side gate-drive signals on the PFC and main switch (oscilloscope-only). If cap ESR is clean but Code 202 persists with primary-rail droop in the telemetry, the PFC stage gate-drive timing or the main switching MOSFET may be marginal. Requires an isolated oscilloscope and AC-mains-rated probes - skill gate, not for casual DIY. If you have the gear, scope the PFC gate at the MOSFET, watch for clean square-wave drive at the PWM frequency, and look for missed cycles or duty-cycle compression under load. Marginal gate drive is rare but real on aged PSUs.

Cross-test against a known-good PSU of the same family. If primary-side rework is beyond your gear, the simplest definitive test is a known-good PSU swap. Pull a healthy P21 / P221 / P222 from another miner, install in the faulty chassis, boot, observe. Code 202 clears with known-good PSU = original PSU is the fault, replace or bench-repair. 202 persists with known-good PSU = chassis-side wiring or AC infrastructure is the fault, back to Tier 2. Match PSU family precisely - cross-stack mixing introduces its own faults.

Stop DIY when Tiers 1-3 are clean but Code 202 still latches under load. The fault is most likely a marginal PSU that needs bench reproduction against a programmable AC source - normal mains supply doesn't reproduce the conditions reliably. D-Central's bench process feeds the PSU AC at controlled voltages from 190V-270V, controlled temperatures, controlled load profiles, and reproduces the trip definitively. Home gear cannot match this. Ship the PSU (with the chassis if practical) and we pin the fault category in one afternoon instead of a weekend of chasing.

Stop DIY when you see scorched silicon, blown caps, or burned silkscreen. Visible damage on the PSU primary side - burned MOSFETs on the PFC stage, exploded caps with electrolyte sprayed across the board, scorched silkscreen near the input filter - is cascade damage. Replacing only the visibly-failed part leaves the unit vulnerable to the next event because adjacent parts are on the same aging curve. D-Central's bench process recaps the full input chain, replaces damaged switching components, inspects the PFC stage for collateral damage, and burns the unit in for 24 hours at full nameplate before return.

Discuss repair-vs-replace honestly. A brand-new MicroBT P21 runs CAD $220-320, P221 $340-420, P222 $420-540 at D-Central; graded-salvage units run roughly 60% of those numbers. A bench-level primary-side rework runs CAD $80-180 depending on what's failed. If the PSU is 5+ years old and you have already had one prior Code 202 event, replacement is usually cheaper long-term than chasing repeat repairs. D-Central quotes up front with photo-documented diagnostic so you can choose. For miners under 3 years repair almost always wins; over 5 years replacement usually does.

Ship with full context and photos. Pack the PSU (or chassis if shipping the whole unit), copies of: your last btminer-api status JSON capture, your WhatsMinerTool PSU telemetry log, your AC-input voltage log if you have one, intake-ambient log, service history (recent moves / swaps / firmware updates / electrical events). Match chassis serial to PSU serial in your ship note - both are on metal plates, don't guess. Every minute of context saved is a minute of turnaround shortened. Canada-wide standard shipping; US / international welcomed. Book a repair slot at d-central.tech/services/asic-repair/ - turnaround 3-7 business days for isolated Code 202.