APW9+ Output Voltage Too Low On Boot

Confirm the symptom with a multimeter on the DC range. Probe at the PSU-to-miner load connector while the rig is mining at full power. Healthy `APW9+` sits at `14.5-15.0 V` sustained; below `13.0 V` is the trip threshold for hashboard PMIC undervoltage protection. Log three cold-start cycles and capture the rail trace at t=0, t=5 min, and t=15 min. If the rail is in spec, this is not the page you need.

Measure AC line voltage at the panel under full load. On 240 V split-phase expect `235-245 V`; on 208 V commercial expect `202-212 V`; on 120 V residential expect `115-125 V`. Loaded readings 10-15 V below nominal mean the circuit is shared with another high-draw appliance (HVAC, dryer, kiln) or the wiring run is undersized. Move the miner to a dedicated `240 V` circuit before spending money on PSU parts.

Verify the wall outlet itself. Check connector torque, look for arc-pitting or browning at the receptacle, and confirm the breaker is sized for the continuous load (`APW9+` at full draw wants a 20 A 240 V circuit minimum, 30 A is the home-miner default). A cooked outlet drops 5-10 V at high current and presents identically to a tired PSU.

Drop the miner profile by 10% and re-test. Lower the frequency or watt limit on the S17 Pro / T17 from stock, restart, and re-measure the PSU rail at the new (lower) load. If the rail recovers to `14.5 V+`, the caps are marginal but not yet dead — undervolting buys you 6-18 months while you plan a re-cap. If the rail still sags, the problem is upstream of load.

Tighten and clean the PSU output load connector. With the rig powered off and bled down, unscrew, inspect contacts for corrosion or arc-pitting, clean with isopropyl alcohol, and reseat firmly. Loose high-current connectors drop voltage at the contact and present as low PSU output even when the rail is fine inside the chassis.

Vacuum and inspect the PSU intake/exhaust grilles. Three to five years of basement dust on a continuous-duty `APW9+` clogs the airflow, drives internal temperature up, and accelerates cap aging. Shop-vac the grilles, blow compressed air through the chassis, and verify the fan spins freely without bearing rumble. Free intervention, mandatory before opening the PSU.

Capture a thermal image of the PSU under load. A `FLIR ONE Pro` clipped to your phone (or any IR thermometer) tells you whether the bulk primary caps are running materially hotter than the rest of the PSU. Bulk caps glowing above `75 °C` while the heatsink and FETs sit at `45-55 °C` is the visual fingerprint of cap-driven low-rail failure. If everything is uniformly hot, the load is too high or airflow is restricted.

Inspect the PSU fan for bearing wear. With the rig powered off, manually spin the fan blade. Should spin freely with minimal noise. Grinding, wobble, or audible bearing rumble = fan replacement before tackling caps. A 60 mm 12 V fan is `$8-$15 CAD` and is the cheapest possible intervention on a hot-running PSU.

Add external cooling on the PSU heatsink as a Tier-1 holdover while you plan the rebuild. A `120 mm` PC fan blowing across the heatsink fins, powered from a USB or 12 V tap, drops internal temperature `5-15 °C` and frequently restores rail voltage on a marginal `APW9+`. Cost `$15-$30 CAD`. Buys 6-12 months while you order parts.

Open the PSU casing and discharge the bulk caps before you touch anything inside. Wait `5 minutes` minimum after power-off and unplug. Use a `10 kΩ 5 W` resistor across each bulk cap's terminals to bleed residual voltage. Verify zero volts with a multimeter before reaching in. The primary rail on a bridge-rectified PFC stage running on `240 V` mains sits at roughly `385 V DC` — this is the step where people who skip safety lose fingertips. Take it seriously.

ESR-test every electrolytic. Use a `Peak ESR70` or equivalent meter on each cap with the PSU unpowered and discharged. Bulk primary caps should read below `0.2 Ω`; secondary low-ESR filter caps should read below `0.1 Ω`. Anything above `0.5 Ω` on a primary cap or above `0.3 Ω` on a secondary is a confirmed replacement. Bulging, vented, or leaking electrolyte is a replacement regardless of ESR.

Source replacement bulk caps to the original spec — common `APW9+` configuration is `2 × 470 µF 450 V` (or `2 × 560 µF 450 V`) low-ESR aluminum electrolytic, `105 °C` rated. Replace with `5000-hour 105 °C` parts: `Nichicon LGW`, `Rubycon BXC`, or `Panasonic EE` series. Don't undersize, don't go cheaper than spec. Verify exact value off the can on your unit before ordering — production runs varied.

De-solder the old bulk caps. Through-hole, solder mass is significant. Temperature-controlled iron at `380 °C` plus solder wick or a desoldering pump. Mark polarity before pulling. Pull straight up to avoid lifting pads. Inspect the pads after extraction — if a pad is lifted, this PSU is now a Tier-4 ship-to-D-Central job; do not try to scab-jumper a primary-rail pad on a 385 V DC supply.

Solder in fresh caps and verify polarity. Polarity matters absolutely on the primary rail — a reversed electrolytic on `385 V DC` vents catastrophically the moment power is applied. Triple-check before energizing. Trim leads. Inspect every joint under `10x` magnification for cold solder, bridging, or whisker. The rebuild's reliability is set by these joints; this is not the step to rush.

Replace the secondary-side `12 V` output filter caps while you're in there. They have aged in lockstep with the bulk caps and are usually the proximate cause of the rail droop you're seeing — primary caps set the filter ripple ceiling, secondary caps set the rail steadiness under transient load. Specify low-ESR `105 °C 5000-hour` parts. Adds `$10-$20 CAD` to the bill of materials and is the difference between a rebuild that lasts 6 months and one that lasts 3 years.

Bring the rebuilt PSU up on a current-limited bench supply or a `100 W` incandescent bulb in series with the AC line for the first power-up. If anything is wrong (reversed cap, solder bridge, lifted pad), the bulb glows bright and the PSU shows zero output instead of venting a cap into your face. After 5 minutes of clean idle, remove the limiter and verify rail voltage at no-load before reconnecting to the miner.

Verify the rebuilt rail under load. Reconnect the miner, start mining at stock profile, and probe the PSU output for `30 minutes` of continuous full-power operation. Healthy rebuild: `14.5-15.0 V` sustained with `<100 mV` ripple. If you have a scope, capture the ripple — anything above `200 mV peak-to-peak` means a secondary cap is still bad or the PWM feedback loop is unhappy.

Stop DIY and book a D-Central PSU bench when: re-cap did not restore the rail, you measure shorts on the switching FET / SR diode / opto-feedback chain, scorch marks or burnt smell are present, the PWM controller IC reads abnormal, or pads lifted during extraction. The bench has the schematic, the test fixtures, and a stock of salvaged-grade donor parts. Bench rebuild `$95-$185 CAD` plus parts; turnaround `5-10 business days`.

D-Central bench process: full PSU strip-down, ESR sweep on every electrolytic, FET / diode / opto verification, PWM controller IC sanity check, full cap replacement with `5000-hour 105 °C` parts, post-repair `4-hour` burn-in at full nameplate load on a programmable test fixture. We document the rail trace before and after — you get the data, not just a green light.

The retire-or-rebuild call. Worth rebuilding if (a) you have multiple `APW9+` units on hand, (b) functional `S17` / `T17`-family miners they're paired with, and (c) rebuild cost is under 50% of the going used-PSU price. Single tired `APW9+` paired with a single `S17 Pro` often points to a step-up to a fresh `APW12` (which handles the higher-draw `S19` / `S21`-class miners you'll buy next anyway). Fleet operators almost always come out ahead on the rebuild; single-rig home miners sometimes don't. Run the math before you order parts.