Volcminer D1 Hydro Coolant Leak and Water Cooling Issues

Kill power at the breaker IMMEDIATELY. Not the web-UI shutdown, not the chassis power switch — walk to the panel and drop the breaker feeding the miner's outlet. Confirm fans stop and the PSU LEDs go dark. Every minute energized on a leaking hydro miner is increased risk of dielectric breakdown, dead-short event, and chassis fire. Web-UI shutdown leaves the PSU standby rail energized; that is not enough.

Contain the spread with absorbent material — old towels, paper towels, oil-absorbent mats, kitty litter — whatever you have on hand. Get fluid off the chassis exterior, off the rack, off the floor, and out from under any rack-neighbour miners. Do not use a household vacuum on liquid; use a wet/dry shop-vac rated for liquid pickup, or absorbent material only.

Photograph everything before you touch it. Wide shots of the rack, close-ups of the chassis, close-ups of any visible leak path, photo of the breaker panel showing power is off. If this becomes a warranty or insurance claim, the photos are the entire case. Capture the chassis serial-number plate and your firmware version too.

Sensory check. If you smell burning plastic, ozone, or hot solder, hear arcing or popping, or see smoke — leave the room, confirm power is off at the panel, and call your municipal fire department non-emergency line for guidance. A wet ASIC that has been energized can keep smoldering for hours. Do not be a hero on this one.

Wait 30 minutes minimum after shutdown for the chassis to cool before opening it. Hydro chassis hold heat. Coolant in a hot loop can flash to vapor when you crack the chassis seal — patience here is insurance against a steam burn.

PPE up before any further work: nitrile gloves and safety glasses, well-ventilated work area. Coolant is generally non-toxic but is a skin and eye irritant; some operators react to propylene glycol. Nitrile is non-negotiable. Latex is not adequate.

Drain the loop into a sealed, labeled container. Open the lowest drain point on the chassis or disconnect the lowest fitting; catch every drop in a clean container with a sealing lid. Do not pour propylene-glycol coolant down a drain. Label the container with date, miner serial, and `USED COOLANT - PROPYLENE GLYCOL`; check your municipal hazardous-waste schedule for disposal.

Inspect every external fitting and visible hose section. Look for cracked fittings, extruded O-rings, swollen or hardened hose, dried-glycol residue indicating a prior weep, and damaged threads. Photograph anything suspect. Replace any fitting or O-ring that fails visual inspection — cost is `<$15 CAD`, the cost of skipping is the chassis.

Replace the failed external fitting or O-ring with a verified-compatible part. Stock VolcMiner spec preferred. If substituting, match the thread standard (likely `G1/4` BSPP or industrial QD — verify on your unit) and the elastomer compatibility (EPDM is compatible with propylene glycol; long-service nitrile is not). Wrench-tight to spec, NOT gorilla-tight — over-tightening cracks manifolds and is a leading cause of `I fixed it and now it leaks worse`.

Refill with manufacturer-spec coolant. Stock VolcMiner coolant if you have it; otherwise an electronics-grade propylene-glycol blend with corrosion inhibitors compatible with aluminum cold plates and brass fittings. Do NOT use automotive antifreeze without verifying inhibitor chemistry (silicates and dyes foul electronics loops). Do NOT use tap water (mineral scale). Do NOT use distilled water alone (no inhibitor package — goes ionic on a powered loop).

Leak-test BEFORE re-energizing. Pressurize the loop manually or per VolcMiner's documented procedure for your hardware revision. Hold pressure for 2 hours. Wipe each fitting dry, lay clean paper towel under each fitting, leave the unit for 2 hours. Any wet spot on the towel = leak still present, return to step 8.

Open the chassis only if you have worked on hydro miners before. A D1 Hydro internal is not the place to learn hydro plumbing. The cold-plate manifold uses precision compression fittings and gaskets that will leak again if reassembled imperfectly. If you have not personally serviced a hydro Whatsminer / Antminer S19 Hydro / NerdQAxe Hydro, skip directly to Tier 4.

If you do open it: photograph orientation and torque-pattern markings before disassembly. Inspect cold-plate fasteners, manifold gaskets, hose-clamp positions, and pump-body seal areas. Look for visible gasket extrusion, hashboard PCB staining, white-powder corrosion bloom (depleted inhibitor), or PCB solder-mask discoloration (voltage stress under wet conditions).

Test any wet hashboard before reassembly. Visual: solder-mask staining, exposed-copper greening, residue on chip packages. Electrical (advanced bench): isolation resistance from each power rail to chassis ground; healthy = megohms, contaminated = kilohms. A board reading kilohms is contaminated and needs ultrasonic clean + bake-out before reuse, or replacement. This is bench territory; if you do not have an isolation tester, ship.

If you reassemble: full leak-test before any power application. Pressurize, hold 2h, inspect. Then run only the pump circuit (no hashing) for 1 hour and re-inspect. Then hash at `50% frequency` for 4 hours and re-inspect. Then full nameplate for 24 hours continuous monitoring. Skipping any of those steps and going straight to nameplate is how a recovered chassis becomes a write-off.

Stop DIY and ship to D-Central whenever ANY of: internal leak event (coolant inside the chassis), wet hashboards, suspected pump-seal failure, manifold gasket failure, GFCI tripped during the leak event, any sign of arcing or burning, second leak after a Tier 2 repair, or unknown coolant chemistry. The math is simple: a `$2500-4000 CAD` D1 Hydro chassis is not worth losing because the operator wanted to skip a `$650-1200 CAD` bench repair.

D-Central bench process for hydro leaks: full chassis disassembly, hashboard isolation testing on calibrated bench equipment, ultrasonic clean of any wet PCB, controlled bake-out of recovered boards, leak-tested manifold reassembly with new gaskets, full pressure-and-flow validation, 24-hour nameplate burn-in before return shipping. We treat hydro repairs as electrical-safety repairs, not `clean it and send it back`.

Ship safely with hydro-specific rules: drain the loop completely, ship empty, flag the drain port. Bag any wet hashboards in anti-static bags with desiccant. Double-box with `≥10 cm` foam on every side (hydro chassis are heavy — shipping damage is real). Include an incident note: leak first noticed when, runtime with leak, exact actions before power-off, photos of leak point and damage, firmware version, loop fill date. Detail saves diagnostic time and lowers your invoice.