Prevention is cheaper than repair
Use a simple schedule: inspect cables and temperatures weekly, clean according to the environment, and check logs after any outage or profile change. Bitmain recommends cleaning miners regularly and stresses that the device must be shut down before cleaning.
The interval depends on the room. A clean data center allows a longer interval than a garage, basement or room with textiles and pets. Do not follow the calendar alone: rising temperatures and falling hashrate are reasons to inspect earlier.
- fully shut down before cleaning
- record the date and result
- check airflow on both sides
- do not clean a running ASIC with compressed air
Intake temperature matters more than room temperature
Manufacturers distinguish intake air from board or chip readings. Limits differ by Antminer series, so a universal forum number does not replace the documentation for the exact model.
Hot exhaust can return to the intake even when the room feels cool near the door. Separate intake and exhaust, leave clearance around the units and test temperatures under sustained load.
- watch inlet and PCB/chip temperatures
- do not confuse room temperature with intake temperature
- check for exhaust recirculation
- reduce load and restore cooling during overheating
Dust, fans and humidity
Dust on heatsinks reduces heat transfer, while dirty fans lose airflow. During cleaning, do not force fan blades at high speed with compressed air and do not allow compressor moisture inside the machine.
Liquid near an ASIC is a separate risk. If a device was cleaned with moisture, it must be fully dry before power is applied. Watch for hot-plastic smells, fan vibration, unusual noise and temperature differences between hashboards.
- clean heatsinks and fan grilles
- avoid condensation and high humidity
- replace a fan with play, grinding or unstable RPM
- never power a wet device
The kernel log can show a problem before the dashboard does
The ASIC web interface usually exposes current and historical logs. They can reveal a missing hashboard, missing chips, network failure, power failure, fan issues or over-temperature protection. Check logs before the miner stops completely: repeated short errors can explain unstable hashrate.
One error does not always mean a failed board. Check control cables, grounding, PSU, temperature and event repetition. If the problem follows a cable or PSU between boards, that is useful diagnostic evidence.
- missing hashboard — check cable and power
- overtemp — check dust, intake and exhaust
- network failure — check Ethernet, DNS and latency
- power failure — check contact, PSU and protection
Post-maintenance checklist
After cleaning, do not restart without a short inspection. Check connectors, fan movement, fasteners and foreign objects. Apply power only after the unit is fully dry and the airflow path is clear.
After startup, compare each board's hashrate, temperature, fan speed, accepted shares and power with the previous record. Check the historical log again after several hours because some issues appear only after the machine warms up.
- take a photo and record the baseline
- check every hashboard after startup
- compare temperature and power
- save logs if an error repeats
| Symptom | First checks | Do not |
|---|---|---|
| Hashrate drops | Temperature, hashboards, kernel log | Raise power limit blindly |
| High temperature | Dust, intake, exhaust, fans | Leave the ASIC in a closed cabinet |
| Missing hashboard | Cable, PSU, contact and grounding | Replace a board without diagnosis |
| Fan error | Connector, RPM and mechanical noise | Block a fan while running |
| Network failure | Cable, DHCP/DNS and latency | Expose the web panel to the internet |
Real photos and source material
These photographs and diagrams are sourced from public archives. Open the source link to check the author and license.
Want to check the numbers for your ASIC right away?
Open calculator →
.jpg?width=1200)
