Avalon 1446 – Temperature Too High

Pull the power cord, wait 60 seconds for the bulk capacitors to drain, then restart. A full cold-boot clears any wedged firmware state around the OVER_TEMP trip flag - MM3.1 occasionally holds the fault across a soft reboot on the 1446. Count to 60 seconds; do not just use the dashboard restart. This step alone clears roughly one in five events on a 1446 that saw a transient thermal spike (neighbour's appliance on the same HVAC circuit, afternoon sun on the mining shed, warm intake air for 10 minutes). If it boots clean and runs a full hour without re-tripping, you caught a transient.

Vacuum the front intake grille and rear exhaust vent with a shop-vac using the soft-brush attachment. Pay particular attention to the four HA1250H12SB-Z fan blades and the heatsink fin packs visible through the grille. Dust-packed fins can raise the heatsink-to-ambient delta by 8-15 C, which is the entire margin between running fine and OT on a warm day. The 1446 dissipates more heat than a 1346 at equivalent ambient, so dust accumulation bites sooner - expect monthly vacuuming in a dusty install, quarterly in a clean one.

Measure intake-air temperature with an IR thermometer or probe thermometer held 5 cm from the front grille during hashing. Target 30 C or below for a Canadian basement or garage install; 35 C is the Canaan absolute maximum per spec. If intake reads above 30 C, open a window, crack the garage door, or move the miner's intake to a cooler corner before touching the miner. The 1446 is particularly unforgiving of high ambient because the A3206 silicon runs hotter than the A3205 at equivalent clocks.

Confirm physical clearance: at least 30 cm in front of the intake, at least 15 cm behind the exhaust. A 1446 on a shelf with no breathing room recirculates its own exhaust - which is exactly how a miner that ran fine last month starts tripping OT this month after someone stacked a storage bin in front of it. Pull everything back from the miner. Clearance is free; use it. Restart and watch for a full hour before declaring the problem solved.

If you run multiple miners in a shed or garage, re-verify airflow direction. 1446s facing each other's intakes cook each other fast. Correct layout is cold-air plenum feeding the front of every miner, hot exhaust ducted outside or into a heat-utilisation channel for home heating. Recirculating the miner's own exhaust into its intake is the single fastest way to hit OT in any install, and the 1446's higher-power thermal envelope makes this failure mode bite harder than it does on a 1346.

Open the chassis, remove each of the four HA1250H12SB-Z fans in turn, and spin them by hand. A healthy dual-ball-bearing fan spins freely for 2-3 seconds after a flick. A failing fan grinds, buzzes, or stops immediately. Replace any fan that fails this test with a drop-in HA1250H12SB-Z. The 1446 uses the same fan as the 1346, 1366, 1466, and 1566 - one spare SKU covers the whole A13/A14/A15 generation. Parts available from Zeus Mining, bit2miner (DF1205012B2 cross-reference for later revisions), or as part of a D-Central parts order.

Measure the 12 V fan rail at the PSU-to-control-board connector under load. Expect 11.8 V to 12.2 V sustained. Below 11.6 V means a tired PSU sagging the fan rail - fans slow, chips heat, miner trips OT, and you chase a thermal fault that is actually a power fault. Swap to a known-good A14/A15-family PSU before assuming the miner is the problem. Critical: 1446 PSUs are NOT cross-compatible with 1346 or 1166 Pro PSUs - control signalling changed between the A13 and A14/A15 generations. Use the correct 1446-spec PSU.

Re-torque the heatsink mounting clips on each of the three hashboards. Canaan's 1446 heatsink clip loosens over 12-18 months of thermal cycling, and the 1446's higher operating temperature accelerates the cycle. A heatsink with even 0.1 mm of lift loses paste contact and the chip underneath cooks while its neighbours run fine. Remove the heatsink, wipe off the old paste with 99% IPA, apply a thin uniform layer of Arctic MX-6 or Thermal Grizzly Kryonaut, reseat with even clip pressure. Do one hashboard at a time so clips do not get mixed between boards.

Install a foam pre-filter on the intake if your install does not already have one. D-Central strongly recommends a washable foam pre-filter on every 1446 install - it catches dust before it reaches the heatsinks, and you can vacuum or rinse it in seconds instead of opening the chassis. On the 1446 specifically, the slightly denser fin pack (vs the 1346) means dust bridges more quickly - a pre-filter pays for itself in months rather than the year-plus timeline on a 1346.

Verify line voltage at the outlet under load. On 240 V split-phase (the correct feed for the 1446's 3420 W draw) expect 235-245 V. On 208 V commercial expect 202-212 V. Low line voltage forces the PSU to pull more current, which heats the PSU internals, which can trip the PSU thermal sensor independently of the hashboard sensors. A 1446 on a stretched 120 V circuit cascades every failure mode in this document - fix the feed before continuing. That is an electrician or panel problem before it is a miner problem.

Snapshot cgminer-api estats output to a text file before any further changes. Run `echo -n '{"command":"estats"}' | nc 127.0.0.1 4028` from the miner or a network-connected laptop and save as 1446-pre-fix.txt. If the miner ends up shipping to D-Central, that pre-fix snapshot is exactly what the bench tech wants. The 1446 is new enough that community repair data is still thin, so your snapshot also contributes to the evolving knowledge base.

Remove a hashboard, strip the heatsinks, and reapply thermal paste on every ASIC on the chain that dominated your PVT_T outlier list. Use Arctic MX-6 or Thermal Grizzly Kryonaut in a thin uniform film - the grain-of-rice heuristic is for CPUs, not a 120-chip hashboard; you want uniform coverage across every chip top. Replace any thermal pad under the PCH or voltage-domain ICs if pads are crumbled, discoloured, or compressed. Budget 60-75 minutes per hashboard on a 1446 the first time, 30-40 minutes once practised.

If one specific chip position is the outlier and paste refresh did not resolve it, reflow that single chip. Preheat the hashboard from below at 150 C for 3 minutes, hot-air from above at 310-330 C for approximately 30 seconds with flux around the chip periphery, natural cool-down. The A3206 BGA tolerates a single reflow cycle reliably. A second reflow on the same chip within 90 days rarely sticks - at that point the chip is dying and needs replacement from graded-salvage inventory.

Flash a known-good MM3.1 firmware build from Canaan's firmware portal at avalonminer.org/firmware-document/. The portal is login-gated and release notes are sparse, but it is the only official source for MM3.1 builds. Verify your hardware revision against the firmware compatibility table before flashing - the wrong MM3.1 for a late-rev 1446 can brick the control board. Flash via the dashboard firmware page over a stable wired connection; never flash over wireless, never flash while hashing, never flash while an OT fault is active. Clear the fault first.

If the fault cleared after firmware update but returned within a week, downgrade to the previous stable build and note the revision numbers. A regression in Canaan's OT threshold logic between builds is a documented failure mode on the A13 generation that carries into MM3.1 on the 1446. Document your working build and flash date in a text file taped to the chassis. The community is still mapping stable-build ranges for the 1446 - your notes help the next operator.

Reseat the control-board ribbon connectors on all three hashboards. The same IDC-style ribbons used on the 1166 Pro and 1246 evolved into near-identical connectors on the 1446 MM3.1, and they oxidize the same way in humid environments. Pull each ribbon fully, wipe contacts with 99% IPA on a lint-free wipe, reseat until the latch clicks. Oxidized ribbons can break the temperature telemetry path specifically, causing a chain to appear cool while running hot - tripping a different sensor and confusing diagnosis.

Stop DIY and ship to D-Central when any of the following are true: every fan replaced, every heatsink cleaned, every chip repasted and the miner still trips OT within 24 hours; reflowed a chip once and OT returned within 30 days; PVT_T[] returns impossible values (negative, above 200 C, or identical across all 120 chips); same chip position trips on two different 1446 units; visible discoloration, burnt smell, or capacitor damage anywhere on a hashboard or the MM3.1. Book a D-Central ASIC Repair slot. Turnaround 5-10 business days, Canada-wide shipping, US and international welcomed.

Pack for shipping: anti-static bags on each hashboard and the MM3.1 separately, double-box with at least 5 cm of foam on every side, include a printed diagnostic note with the kern.log excerpt (actual OT lines), MM3.1 firmware build, PSU model (confirm A14/A15 generation), measured line voltage, intake ambient, fan tach readings before failure, and every Tier 1-3 step you have already run. The bench tech starts from your notes - better notes, faster and cheaper repair. Never ship hashboards loose or in single-box packaging; A3206 BGAs do not survive careless handling.