Crypto Mining Blog
Top 10 Mistakes Beginners Make When Buying ASIC Miners (and How to Avoid Them)
TL;DR
-
Don’t buy on TH/s alone—J/TH (efficiency) is the margin maker.
-
Model power cost, difficulty paths, and uptime, not just today’s “profitability.”
-
Verify all-in price (rig + shipping + GST/duties + PSUs + cables + cooling).
-
Avoid scam or gray-market listings; pick reputable sellers and clear warranties.
-
Plan for noise/heat/power at your actual site before you pay.
-
Start modular, keep spares, and have a maintenance plan.
1) Chasing TH/s and ignoring J/TH
Why it hurts: High hashrate with poor efficiency burns cash as difficulty rises and power prices fluctuate.
Fix: Sort candidates by J/TH, then compare capex. Efficient rigs keep margins longer.
2) Trusting “profit per day” screenshots
Why it hurts: Those snapshots assume today’s price/fees/difficulty and ignore your tariff.
Fix: Run scenarios: flat, +5%, +10% difficulty per retarget, two BTC price paths, your ¢/kWh.
Use our Bitcoin Mining Calculator → /pages/bitcoin-mining-calculator.
3) Forgetting the power bill details
Why it hurts: Demand charges, peak TOU windows, and network fees can blow out costs.
Fix: Know ¢/kWh + $/kW demand and TOU windows. Add 10% aux (fans/pumps) unless measured.
Target PUE: ≤ 1.15 (air) or ≤ 1.10 (immersion).
4) Buying before checking noise & cooling
Why it hurts: Domestic sites get neighbor complaints or thermal throttling/outages.
Fix:
-
Air: straight-through ducting, baffles, quality filters, variable-speed fans.
-
Immersion: budget tank + fluid + dry cooler and fluid handling.
Aim < 50–60 dBA at boundary (local rules vary).
5) Ignoring total landed cost
Why it hurts: Shipping, insurance, GST/duties, PSUs, cabling, racks, and breakers add up.
Fix: Request a line-item quote. Add 10–15% contingency. Confirm lead times.
6) Buying from unknown or “too-cheap” sellers
Why it hurts: Counterfeit PSUs, used gear passed as new, or no recourse on DOA units.
Fix: Buy from reputable vendors (invoices, serials, warranty terms). Search reviews, and prefer escrow or secure payment options.
7) Not standardizing the fleet
Why it hurts: Mixed models = more spares/firmware headaches and lower uptime.
Fix: Pick one or two models you can maintain well; stock fans/PSUs/hashboard spares.
8) Skipping site power checks
Why it hurts: Tripped breakers, unbalanced phases, and hot panels kill uptime.
Fix: Verify service amps/phase, breaker sizes, cable gauge, and phase balance. Add a site kW cap in your EMS and staggered boot.
9) Underestimating maintenance
Why it hurts: Dust, failing fans, and loose lugs cost more than parts—they cost downtime.
Fix: Set a maintenance cadence (weekly dust/filter, monthly thermal scan, quarterly health audit). Track uptime, PUE, peak kW.
10) Overcommitting before proving unit economics
Why it hurts: Locking in too many units ties you to a bad tariff or hot site.
Fix: Start modular (2–3 rigs), prove ROI for 30–90 days, then scale in repeatable blocks.
Quick Buyer’s Checklist (printable)
-
Tariff mapped: ¢/kWh, $/kW demand, TOU windows
-
Calculator run with multiple difficulty/price paths
-
Spec sheet: TH/s, J/TH, W, noise, PSU included?
-
Site ready: power, breakers, PUE target, cooling plan
-
Total landed cost: rig + PSU + shipping + GST/duty + cables + racks
-
Vendor vetted: invoice, serials, warranty, RMA process
-
Spares: fans, PSUs, one known-good hashboard
-
EMS: ramp sequencing, site kW cap, pool failover
-
Maintenance plan & logs
Comparison Priorities (what to sort by)
-
J/TH (efficiency)
-
Capex per TH (after landing costs)
-
Reliability & noise (site specific)
-
Cooling fit (air vs immersion readiness)
-
Warranty & vendor support
FAQ
Are used miners worth it?
Sometimes—if efficiency still pencils with your power price and you budget for spares. Avoid heavily overclocked or corroded units.
How many rigs should a beginner start with?
Enough to learn maintenance without risking the farm—often 1–3 rigs. Scale in blocks once KPIs look good.
Hosting vs self-run?
Hosting reduces buildout but adds counterparty risk and fees. Read SLAs, uptime credits, and curtailment revenue split carefully.
