Board-level diagnosis and repair, Bitmain machines first. Send units to the Austin bench in any quantity, or bring our technicians to your site. Downtime is the real cost, so most units go back in the rack within 48 hours.
Ship boards and units to the bench, or have the bench come to you. Which one makes sense comes down to how many machines you run.
Send whole miners, hashboards, control boards or power supplies to our Austin bench. One unit or a pallet — there is no minimum. Every unit is diagnosed, quoted, repaired on approval and load tested before it ships back.
For sites running 1,000 machines or more we bring the bench to you: technicians, tools and test fixtures on your floor, so boards never leave the building. Scope, duration and staffing differ at every site, so this is quoted case by case.
Antminer is what our bench does day in, day out — S17 through S21, hashboards, control boards and chip-level work. MicroBT and Canaan units are taken on request.
APW9, APW12 and APW17 repair is worth doing at volume, not one at a time. Send them as a batch and we rail test, replace capacitors and load test the lot.
Bench or on site, every unit leaves with a written record of what failed and what was replaced.
Tell us the model, the fault and how many units. We reply with a drop-off slot or a shipping label, and a quote before any work starts.
Send to the Austin bench with the RMA we email you. No minimum quantity.
Technicians come to your facility. Sites of 1,000 machines and up, quoted case by case.
48 hours average from arrival, longer for chip-level work.
Fault maps, reflow profiles and the stock firmware images we use on the bench. Ask for a document by name and we email the file back the same business day.
Which chip a hashboard error code points at
Temperatures and dwell times per chip
Which PSU letter may replace which
Bitmain and MicroBT, with flashing steps
What we sign a repaired board off against