How to Remove the Power Supply from an All-in-One Mining Machine

Start with safety and logs

Power down before opening a miner, label cables before moving boards, and capture logs before repeated reboots erase useful evidence. Record model, firmware, pool, uptime, fan speed, temperature, reject rate, chain count, and the exact error text.

Confirm the fault class

Separate configuration faults from hardware faults first. Pool errors, DNS failures, bad worker names, overheating, weak power, fan faults, and missing hashboards can look similar from the dashboard but require different fixes.

Document the test path

Change one variable at a time and keep the before/after result. Note cable swaps, PSU swaps, firmware changes, pool changes, fan replacements, ambient temperature, and whether the fault follows a hashboard, control board, network, or power source.

When to escalate

Escalate to professional repair when there is a burned smell, melted connector, breaker trip, corrosion, repeated hashboard loss, liquid exposure, or a board-level fault that returns after a basic cable, power, firmware, and airflow check.

After the fix

Run the miner long enough to confirm stable accepted hashrate, fan behavior, chip temperature, reject rate, and pool-side reporting. A dashboard that looks normal for five minutes is not enough evidence for a recurring power, heat, or hashboard fault.