HP DPS-1200QB Breakout Fan Pin Full-RPM Noise Fix

Buy a quality breakout that does the fan-control work for you. Parallel Miner's `DPS-1200FB` / `DPS-1200QB` breakout (`~$45-65 CAD`) includes a thermistor-driven fan-control circuit that bonds the back-edge `FAN_CMD` pin to a temperature-tracking PWM. Drop-in replacement for whatever cheap breakout you have now. No soldering on the PSU side. Worth the upgrade if you are not interested in soldering iron work and want a clean, supported path that tracks heatsink temperature properly under load.

Verify the breakout's thermistor is bonded to the PSU heatsink. The thermistor is useless if it's flapping in free air. With the breakout installed, route the `10 kohm` NTC to the PSU's secondary heatsink and tape it down with `3M 8810` thermal tape (or equivalent). The thermistor must be in physical contact with the heatsink fin material, not the PSU plastic enclosure. Re-check tape adhesion every `12 months` - dried tape drops the thermistor off and reverts the PSU to default-loud behaviour.

Power up and listen. Within `30 seconds` of AC apply, the fan should drop from full-RPM startup transient to a sane idle speed (`~3,000-5,000 RPM`, `~40 dBA` at `1 m`). Apply a load (Bitaxe hashing): fan should track up modestly with heatsink temp. If the fan stays at full RPM after `30 seconds`, the thermistor isn't seated correctly - re-seat it before continuing.

Document the build on the PSU case. Sharpie on the case: PSU PN, breakout vendor and PN, install date, mod path used (1 / 2 / 3), and `OTP` verified trip temp once you've run the verification test. Future-you doesn't have to re-test, and if the rig ever needs warranty / failure analysis, you have the verification record.

If your existing breakout doesn't have fan control and you're not ready to solder: a `120 mm` external fan ducted to blow into the PSU's intake vent is a brute-force quiet path. Run a `120 mm` Noctua at `40%` PWM, point it at the intake, and the PSU's own fan will drop RPM in response to the cooler intake air it sees. Doesn't fix the root cause but gets you `~15 dBA` reduction for `$25 CAD` and zero solder joints.

Tier 2 entry: confirm the fan-command pin per the datasheet for your exact PN. Don't trust generic forum posts - pin numbers vary across `DPS-1200FB-1A` / `DPS-1200FBA` / `DPS-1200QB` / `DPS-750RB-A`. Pull the HP datasheet or a reputable breakout vendor's per-PN pinout PDF. Mark the pin on the back-edge connector with a Sharpie. Mis-routing a `28 AWG` jumper to the wrong signal pin can brick the controller IC.

Build the resistor divider for fixed-duty fan command. For a fixed `~50%` PWM-equivalent voltage on the fan-command pin, a `10 kohm` resistor from `+5 Vsb` (PSU standby rail - verify pin number per datasheet) to the fan-command pin, and a `10 kohm` resistor from the fan-command pin to `COM`, gives you a clean `2.5 V` DC on the command pin. The fan controller reads this as 'moderate command, run mid-RPM.' Adjust the divider ratio (`4.7 kohm` / `10 kohm`) to bias duty up or down.

Solder the divider on the breakout PCB, not on the PSU back-edge. The PSU is the expensive part - keep all rework on the breakout. Use `1/4 W` resistors (current is negligible, thermal headroom is the only reason). Heat-shrink the joints. Route any new wires away from the high-current `+12 V` bus on the breakout to prevent inductive coupling onto the small-signal command line.

Verify the divider voltage with the PSU running. PSU enabled, no load, DMM on the fan-command pin to `COM`. Should read `~2.5 V` (or whatever your divider math computes). Fan should now spin at a fixed mid-RPM (`~6,000-8,000 RPM` on a stock `40 mm` blower) regardless of load. If still at full RPM, the divider isn't injecting voltage - recheck pin assignment. If too quiet, bias the divider up.

Verify safety with the hot-air-gun OTP test. PSU running unloaded, cover off, hot-air gun on low aimed at the secondary heatsink. Watch DMM on `+12 V` output. Output must cleanly drop to `0 V` when heatsink hits `~85 degrees C` (`OTP` engaged) and recover when heatsink falls below `~70 degrees C`. If `OTP` doesn't engage, stop and revert the mod - the PSU is no longer safe to run unattended. Document the verified trip temp on the case.

Tier 3 entry: build a thermistor-driven fan-command circuit if your breakout doesn't have one. `10 kohm` NTC thermistor (Vishay `NTCLE100E3103JB0` or equivalent) thermally bonded to the PSU heatsink. Voltage divider with a `10 kohm` series resistor off `+5 Vsb`. Output of the divider feeds an `LM358` op-amp wired as a comparator with adjustable threshold (`10 kohm` pot on the reference input). Op-amp output drives the fan-command pin via a `1 kohm` series limit resistor. Tune the pot for the fan curve you want.

Replace the stock `40 mm` blower with a `120 mm` Noctua. Drill out the rear blower mount, fab or `3D-print` a `120 mm` adapter shroud (designs available on Thingiverse / Printables - search 'DPS-1200FB 120mm fan adapter'), bolt a Noctua `NF-A12x25` PWM into the rear position. Cut and tape the original `40 mm` tach line. Run the `120 mm` off the breakout's `+12 V` rail through a buck or PWM controller (`50%` duty is typically silent and adequate).

Validate cooling under sustained load. Run a representative load (S9 hashing at `1100 W`, or a benchtop electronic load at `100 A / 12 V`) for `30 minutes`. IR thermometer on the heatsink: should stabilize at `<65 degrees C` with ambient at `~22 degrees C`. If heatsink runs hotter, your `120 mm` is undersized or the duct is leaking - re-fab the shroud and re-test. Don't leave a Path 3 mod under-validated.

Implement intake ducting if PSU is in a tight enclosure. Quiet-modded PSUs depend on having clean intake airflow - the fan is doing less work, so the air it does move has to count. A `25 mm` foam duct from a clean-air source (rather than re-circulating warm cabinet air) drops sustained heatsink temps by `5-10 degrees C` in our bench testing. Worth the cardboard and gaffer tape, especially in compact basement-shelf installations.

Document the full mod stack on the PSU case for future maintenance. Path used (1 / 2 / 3), fan PN if replaced, thermistor PN if added, resistor values, verified `OTP` trip temp, date. Six months from now, when the rig is loud again because something else failed (dust, dried tape, capacitor drift), the documentation tells you what was originally validated and what to recheck first.

Tier 4 - stop DIY. Stop and ship to D-Central when: you burned a fan-controller IC through a wiring error, the PSU has visible scorching or burnt-component smell, secondary regulation is unstable (`+12 V` oscillates `+/- 0.5 V` even with mod fully removed), or you're planning a multi-PSU paralleled rig and want the wiring vetted before powering up. The shop can recover most fan-controller blowouts on `DPS-1200FB` family PSUs at the chip level for less than the cost of a replacement PSU.

What D-Central does at the bench: test fixture with programmable DC load to stress the PSU across its rated range, scope on the fan-command pin to verify quiet-mod behaviour, hot-air-gun verification of `OTP` trip temp, scope on the `+12 V` rail under transient load to confirm regulation. Quiet-mod build review for paralleled multi-PSU S9 rigs (the `I_share` paralleling concern is real - see the sibling `dps-1200fb-breakout-voltage-sense-miswire` page). Free quote, parts at cost, North American return shipping included.

Ship safely. Anti-static bag the PSU and breakout separately, double-box with `5 cm` of foam on every side, include a note with: PN of the PSU, PN of the breakout, mod path attempted, observed symptom, and your contact info. Saves diagnostic time and saves you money on the repair invoice. Book at https://d-central.tech/services/asic-repair/ - free quote, North American return shipping included.