What This Error Means
An “ASIC Chip Error” — also shown as “Dead ASIC,” “Chip [X] not responding,” “ASIC [X] failed,” or displayed as red/missing chips in the miner status grid — means one or more individual ASIC hashing chips on a hashboard have failed. Each Antminer hashboard contains dozens to over a hundred ASIC chips connected in a daisy chain. When a chip dies, it either stops responding entirely or produces hardware errors instead of valid hash computations.
ASIC chips are the core computing elements that perform SHA-256 (or other algorithm) calculations. Each chip failure directly reduces the hashboard’s hashing capacity. A few dead chips reduce performance proportionally, while a dead chip at a critical position in the daisy chain can take down an entire chain or section of the board.
Common Causes
- Thermal stress and degradation — Prolonged operation at high temperatures accelerates electromigration within the chip die, eventually causing internal failure. This is the number one cause of ASIC chip death and is largely preventable with proper cooling.
- Manufacturing defect — Some chip batches have higher failure rates due to silicon quality variations. This is especially noted in certain S17 and T17 production runs.
- Power surge or voltage spike — A sudden voltage spike can destroy the delicate internal structures of an ASIC chip. Lightning events, power grid fluctuations, and PSU failures can all cause voltage spikes.
- Overclocking beyond safe limits — Running chips at voltages and frequencies above their rated specifications dramatically accelerates wear and can cause immediate failure.
- Physical damage — Impact, flexing of the hashboard, or improper handling can crack solder joints between the chip and the PCB, causing intermittent or permanent failure.
- Solder joint failure — Over time, thermal cycling (heating and cooling with each power cycle) can crack the ball grid array (BGA) solder connections between the chip and the board.
Step-by-Step Fix
Important: Dead ASIC chip replacement requires micro-soldering skills and BGA rework equipment. The steps below cover diagnosis and the limited fixes a home miner can attempt. Chip replacement itself is a professional repair procedure.
Step 1: Identify Dead Chips
Access the miner web interface and navigate to the ASIC status or chip status page. This displays a visual grid showing every chip on every hashboard. Dead chips are typically shown as red, X, or blank. Document which chain and which chip position(s) are affected. Take a screenshot for reference.
Step 2: Count and Assess the Damage
Count the total dead chips per hashboard. A few dead chips (1-5) on a hashboard is common on older miners and does not necessarily require immediate action — the board will still hash at reduced capacity. A large number of dead chips (10+) or a rapidly increasing count indicates accelerating failure that will eventually take the board offline entirely.
Step 3: Check Chip Temperatures
Look at the temperature readings for chips near the dead ones. If surrounding chips are running significantly hotter than average, the dead chip may be causing increased thermal load on its neighbors, creating a cascade effect. Improving cooling may slow the progression of chip failures.
Step 4: Reflash Firmware
Sometimes what appears as a dead chip is actually a communication error. Reflash the latest firmware and reboot. After the fresh firmware initializes, check the chip status again. If the “dead” chips now appear healthy, the issue was firmware-related, not hardware.
Step 5: Reseat Hashboard Connections
Power off and reseat all ribbon cables and power connections for the affected hashboard. A poor data connection can cause the control board to misidentify chips as dead when they are actually just not communicating properly.
Step 6: Clean and Improve Cooling
Even if dead chips cannot be revived by a home miner, reducing thermal stress on the remaining chips prevents further failures. Clean all heatsinks thoroughly, ensure optimal airflow, and consider reducing operating frequency slightly if the board is showing multiple chip failures — this extends the remaining lifespan.
Step 7: Monitor Over Time
If the number of dead chips is small and stable, you can continue operating at reduced hashrate while planning for repair. Check chip status weekly. If new chips are dying at a rate of more than 1-2 per month, the board is in progressive failure and should be repaired before more chips are lost.
Advanced Diagnosis
Via SSH, detailed chip diagnostics:
ssh root@[miner-ip]
cat /var/log/messages | grep -i "chip"
cat /var/log/messages | grep -i "asic"
# Look for chip enumeration counts and specific chip failure messages
# Messages like "chip XX nonce error" or "chip XX timeout"Hardware error analysis: A chip that is failing but not fully dead will produce hardware errors (HW errors). In the miner status, check the HW error count and rate. A high HW rate from a specific chain indicates a chip on that chain is producing invalid results. This chip is consuming power and generating heat without contributing valid hashes — it is costing you money.
Voltage domain mapping: On newer Antminer models (S17+, S19 series), hashboards are divided into voltage domains, each powering a group of chips. If all dead chips are within the same voltage domain, the voltage regulator for that domain may be the root cause rather than the chips themselves. Replacing the voltage regulator is often cheaper than replacing multiple chips.
When to Get Professional Help
Seek professional ASIC repair for:
- Any dead chip replacement — this requires BGA rework station, hot air soldering, and replacement chips. Not a DIY repair.
- Progressively increasing chip failures — the root cause needs diagnosis before it takes down the entire board
- An entire voltage domain showing dead chips — voltage regulator repair required
- Chips that appear dead even after firmware reflash and cable reseating
- High hardware error rates from a specific chain
D-Central Technologies performs chip-level ASIC repair on all Antminer hashboards. We use professional BGA rework stations to remove dead chips and solder replacement chips. Our technicians diagnose root causes including voltage regulator failures, thermal damage, and manufacturing defects. Submit a repair request here
Affected Models
ASIC chip failures occur on all miner models. Models with higher documented failure rates include: S17 and S17 Pro (particularly early batches), T17 series, and overclocked S9 units. The S19 series generally has better chip reliability, though the S19 XP’s higher power density makes it susceptible to thermal-related failures. S21 series chips are newer and show lower failure rates so far.
Related Error Codes
- Low Hashrate / Hashrate Drop — Dead chips directly cause proportional hashrate loss
- Chain X Only / Missing Hashboard — Enough dead chips can prevent chain initialization entirely
- Temp Too High — Overheating accelerates chip failure
- Voltage Error — Voltage regulator failure can kill multiple chips simultaneously
- EEPROM Error — Sometimes misdiagnosed as chip errors
Frequently Asked Questions
Can dead ASIC chips be repaired or only replaced?
Dead ASIC chips must be replaced — they cannot be repaired. The chip is a complex integrated circuit that has failed internally. Repair involves desoldering the dead chip using a BGA rework station (hot air at precisely controlled temperatures), cleaning the pads, and soldering a new replacement chip. This is a specialized skill that requires professional equipment.
How many dead chips can a hashboard tolerate?
It depends on the chip position and the miner model. Most hashboards can tolerate a few dead chips with proportional hashrate loss. However, if a dead chip sits at a critical position in the communication daisy chain, it can prevent all downstream chips from being detected, making the impact much larger than a single chip. On some models, a single dead chip in position 0 or the last position can take down the entire chain.
Is it worth repairing a hashboard with many dead chips?
It depends on the miner model and current Bitcoin mining profitability. For newer models (S19 Pro, S21), repairing even 10+ chips is typically worthwhile because the hashboard’s earning potential justifies the repair cost. For older models like the S9, the economics may not favor extensive chip repair. Contact D-Central for a repair-vs-replace analysis specific to your situation.
Can I prevent ASIC chip failures?
You can significantly reduce chip failure rates with proper maintenance: keep chip temperatures below 85C (lower is better), clean heatsinks regularly, use surge protection, avoid overclocking beyond manufacturer specifications, and ensure stable power delivery. Miners running in cool, clean environments with quality power have dramatically lower chip failure rates than those in hot, dusty locations with unstable power.