NerdQAxe++ – Undersized 12V PSU / XT30 Amp Fault

Retire any PSU rated under `10A` / `120W` from NerdQAxe++ service. This is the single most-effective fix for 80% of reports. A `12V` / `12A` (`144W`) unit from Mean Well, Alitove, or a quality Amazon listing costs `$25-$45 CAD` and eliminates the root cause. Don't get by with an `8A` brick because the miner happens to boot - it will kill itself slowly. For NerdQAxe (non-plus) the floor is `5A` / `60W`, with `8A` / `96W` recommended; for Hydro add `1A` for the pump.

Use a PSU with a proper XT30 male pigtail from the manufacturer, or a `12V` barrel supply with a `12V`-to-XT30 adapter from a reputable supplier. Avoid soldering your own adapter unless you can verify the XT30 pins are genuine Amass and the wire is at least `16 AWG` silicone for `10A+`. A weak DIY adapter is the #2 root cause after an undersized PSU.

Route the power cable away from the miner's air intake. PSU heat + miner heat compounding in a closed space is how PSUs sag faster. Keep the PSU on a hard surface with clearance above and below it; a small desk fan pointed at the brick buys you an easy `5-10 C` drop in steady-state temp.

Verify the miner is running at stock before adding overclock. Stock NerdQAxe++ is `~4.8 TH/s` / `~130 W`. If you're not hitting those numbers clean, don't overclock - solve the PSU or cooling issue first. Overclocking a marginal system is how a `10A` PSU becomes a `13A` load in five minutes.

If you own a Kill-A-Watt, leave it inline permanently. `$25 CAD` of continuous wall-draw visibility is worth more than every other monitoring tool on this list. A sudden jump or drop in wall draw is almost always the first sign of a failing PSU, dying chip, or thermal event - and you'll see it days before AxeOS surfaces anything.

Measure the `12V` rail at the XT30 under load. DC multimeter, probe on the board-side XT30 pins while the NerdQAxe is hashing at stock. Expected: `11.9-12.1V` sustained. Anything below `11.6V` = upstream sag. Compare this to PSU-side measurement - if PSU is clean but board is sagging, the connector / wire / adapter is resistive. If PSU itself is sagging, the supply is undersized or dying. Record both numbers before you change anything.

Replace the XT30 pigtail with a genuine Amass part. Genuine Amass XT30 has a yellow polarity key on the female side and a metallic copper contact visible when you look down the barrel. Counterfeits have muddy plating and loose-fitting shells. `16 AWG` silicone wire minimum for the pigtail, `14 AWG` preferred. Solder the wires with flux and proper heat - cold joints are where all the XT30 heat originates. Heat-shrink over the solder, then outer heat-shrink for strain relief.

Cross-check PSU efficiency curve. A PSU rated `85%` efficient at `50%` load may only be `78%` efficient at `95%` load. Running any supply at above `90%` of nameplate continuously = heat + reduced efficiency + faster aging. Derate your PSU choice: size for `60-70%` of its rating at miner full-load. That's why the NerdQAxe++ at `130 W` real draw wants a `180 W` PSU, not a `144 W` one, if you plan to run it 24/7 for years.

Update NerdQAxe firmware to the latest stable. The ESP-Miner fork maintained by shufps for NerdQAxe ships brownout-detection hardening, fan-curve fixes, and `VCORE` telemetry improvements. Use the NerdQAxe Web Flasher via a Chrome / Edge browser, or the D-Central firmware update guide. Do not run pre-release builds on a production NerdQAxe chasing this fix.

Clean the XT30 contacts. Isopropyl `99%` + a small nylon brush. If you've had repeated heating events, the contacts have oxidized and clean metal is buried under a thin insulating layer. Clean both male and female, let dry fully, reconnect. Combined with a re-crimp this clears ~70% of resistive-connector reports.

Scope the `12V` rail at the XT30 under load. DC-coupled oscilloscope, `20 MHz` bandwidth limit, probe on the board-side XT30 pin with ground on the miner chassis. Expected: `11.9-12.1V` DC with under `100 mV` peak-to-peak ripple. Ripple above `300 mV` = the PSU output caps are dying or undersized - the TPS546 VRMs will cover for this up to a point, but sustained noise degrades BM1370 silicon over months. Replace the PSU.

Scope `VCORE` at one of the TPS546 output caps. DC-coupled, short ground lead, probe on a ceramic cap near the BM1370 die. Expected: `1.05-1.15V` DC with under `20 mV` ripple during hashing. If you see large VCORE excursions that track the `12V` ripple from the prior step, the TPS546 is losing its regulation loop because the input rail is too noisy. Again, the real fix is upstream - size the PSU correctly.

Inspect the on-board XT30 socket. Look for discolouration, lifted solder joints on the through-hole pads, or a wobble in the socket. If the socket has been running hot (housing melted, board near-burn), the copper pour underneath may be damaged even if it looks OK - reflow the joints or replace the socket. Use `16 AWG` tinned wire and `60/40` solder; lead-free is fine if your station can deliver the heat.

Check the TPS546 regulators for heat damage. Run the miner for 15 minutes at stock, then IR-scan each TPS546. Healthy: `55-70 C` under load. Suspect: above `85 C` or wildly uneven across the four VRMs. A cooked TPS546 will fault `Device ID mismatch` or silently run out of spec - replace the part from Digi-Key / Mouser (`TPS546D24A` / `D24S`, hot-air rework only, no soldering iron).

If a BM1370 is suspected damaged from sustained undervoltage stress, reflow first (same profile as the Bitaxe / Antminer S21 family: `150 C` bottom-side preheat, `310-330 C` top-side hot air for ~`30 s`). Reflow before replacement because BM1370 chips are not cheap and reflow recovers a surprising number of dead chips whose solder joints were stressed by thermal cycling. If reflow doesn't recover, replace with a graded die.

Stop DIY when: the XT30 socket's on-board pads are lifted or burnt, the `12V` bus copper is discoloured or damaged, a TPS546 has visibly failed with collateral board damage, a BM1370 die is non-responsive after reflow, or you've replaced the PSU and XT30 pigtail twice and the symptoms persist. You're in bench territory. Book D-Central's NerdQAxe repair service.

D-Central bench process: full inspection under magnification, power-path probe with scope (wall -> PSU -> XT30 -> `12V` bus -> each TPS546 -> each BM1370), TPS546 replacement if needed, BM1370 replacement from graded stock if reflow fails, XT30 socket reflow or replacement, post-repair 24-hour burn-in at stock hashrate. We track NerdQAxe failure patterns across our inventory so every repair feeds back into our recommended PSU and XT30 bill-of-materials for future customers.

Ship safely. Anti-static bag, bubble wrap, rigid outer box. Include a one-page note: PSU brand + rating, how long in service, symptoms observed, whether it was running stock or overclock, whether the XT30 was stock / re-crimped / replaced, ambient room temp. Canada-wide and worldwide shipping welcomed - the NerdQAxe community is global and so are our repairs.